Skip navigation.

BI & Warehousing

Data Integration Tips: ODI – One Data Server with several Physical Schemas

Rittman Mead Consulting - Tue, 2014-12-30 16:25

Yes, I’m hijacking the “Data Integration Tips” series of my colleague Michael Rainey (@mRainey) and I have no shame!

DISCLAIMER
This tip is intended for newcomers in the ODI world and is valid with all the versions of ODI. It’s nothing new, it has been posted by other authors on different blogs. But I see so much people struggling with that on the ODI Space on OTN that I wanted to explain it in full details, with all the context and with my own words. So next time I can just post a link to this instead of explaining it from scratch.

The Problem

I’m loading data from a schema to another schema on the same Oracle database but it’s slower than when I write a SQL insert statement manually. The bottle neck of the execution is in the steps from the LKM SQL to SQL. What should I do?

Why does it happen?

Loading Knowledge Modules (LKMs) are used to load the data from one Data Server to another. It usually connects to the source and the target Data Server to execute some steps on each of them. This is required when working with different technologies or different database instances for instance. So if we define two Data Servers to connect to our two database schemas, we will need a LKM.

In this example, I will load a star schema model in HR_DW schema, using the HR schema from the same database as a source. Let’s start with the approach using two Data Servers. Note that here we use directly the database schema to connect to our Data Servers.

Two Data Servers connecting to the same database instance, using directly the database schema to connect.

And here are the definitions of the physical schemas :

Physical Schemas

Let’s build a simple mapping using LOCATIONS, COUNTRIES and REGIONS as source to denormalize it and load it into a single flattened DIM_LOCATIONS table. We will use Left Outer joins to be sure we don’t miss any location even if there is no country or region associated. We will populate LOCATION_SK with a sequence and use an SCD2 IKM.

Mapping - Logical tab

If we check the Physical tab, we can see two different Execution Groups. This mean the Datastores are in two different Data Servers and therefore a LKM is required. Here I used LKM SQL to SQL (Built-In) which is a quite generic one, not particularly designed for Oracle databases. Performances might be better with a technology-specific KM, like LKM Oracle to Oracle Pull (DB Link). By choosing the right KM we can leverage the technology-specific concepts – here the Oracle database links – which often improve performance. But still, we shouldn’t need any database link as everything lies in the same database instance.

Mapping - Physical tab

 

Another issue is that temporary objects needed by the LKM and the IKM are created in the HR_DW schema. These objects are the C$_DIM_LOCATIONS table created by the LKM to bring the data in the Target Data Servers and the I$_DIM_LOCATIONS table created by the IKM to detect when a new row is needed or when a row needs to be updated according to the SCD2 rules. Even though these objects are deleted in the clean-up steps at the end of the mapping execution, it would be better to use another schema for these temporary objects instead of target schema that we want to keep clean.

The Solution

If the source and target Physical Schemas are located on the same Data Server – and the technology can execute code – there is no need for a LKM. So it’s a good idea to try to reuse as much as possible the same Data Server for data coming from the same place. Actually, the Oracle documentation about setting up the topology recommends to create an ODI_TEMP user/schema on any RDBMS and use it to connect.

This time, let’s create only one Data Server with two Physical schemas under it and let’s map it to the existing Logical schemas. Here I will use ODI_STAGING name instead of ODI_TEMP because I’m using the excellent ODI Getting Started virtual machine and it’s already in there.

One Data Server with two Physical Schemas under it

As you can see in the Physical Schema definitions, there is no other password provided to connect with HR or HR_DW directly. At run-time, our agent will only use one connection to ODI_STAGING and execute code through it, even if it needs to populate HR_DW tables. It means that we need to be sure that ODI_STAGING has all the required privileges to do so.

Physical schemas

Here are the privileges I had to grant to ODI_STAGING :

GRANT SELECT on HR.LOCATIONS TO ODI_STAGING;
GRANT SELECT on HR.COUNTRIES TO ODI_STAGING;
GRANT SELECT on HR.REGIONS TO ODI_STAGING;

GRANT SELECT, INSERT, UPDATE, DELETE on HR_DW.DIM_LOCATIONS to ODI_STAGING;
GRANT SELECT on HR_DW.DIM_LOCATIONS_SEQ to ODI_STAGING;

Let’s now open our mapping again and go on the physical tab. We now have only one Execution Group and there is no LKM involved. The code generated is a simple INSERT AS SELECT (IAS) statement, selecting directly from the HR schema and loading into the HR_DW schema without any database link. Data is loaded faster and our first problem is addressed.

Mapping - Physical tab without LKM

Now let’s tackle the second issue we had with temporary objects being created in HR_DW schema. If you scroll upwards to the Physical Schema definitions (or click this link, if you are lazy…) you can see that I used ODI_STAGING as Work Schema in all my Physical Schemas for that Data Server. This way, all the temporary objects are created in ODI_STAGING instead of the source or target schema. Also we are sure that we won’t have any issue with missing privileges, because our agent uses directly ODI_STAGING to connect.

So you can see it has a lot of advantages using a single Data Server when sources come from the same place. We get rid of the LKM and the schema used to connect can also be used as Work Schema so we keep the other schemas clean without any temporary objects.

The only thing you need to remember is to give the right privileges to ODI_STAGING (or ODI_TEMP) on all the objects it needs to handle. If your IKM has a step to gather statistics, you might also want to grant ANALYZE ANY. If you need to truncate a table before loading it, you have two approaches. You can grant DROP ANY table to ODI_STAGING, but this might be a dangerous privilege to give in production. A safer way is to create a stored procedure ODI_TRUNCATE in all the target database schema. This procedure takes a table name as a parameter and truncates that table using the Execute Immediate statement. Then you can grant execute on that procedure to ODI_STAGING and edit your IKM step to execute that procedure instead of using the truncate syntax.

 

That’s it for today, I hope this article can help some people to understand the reason of that Oracle recommendation and how to implement it. Stay tuned on this blog and on Twitter (@rittmanmead, @mRainey, @markrittman, @JeromeFr, …) for more tips about Data Integration!

Categories: BI & Warehousing

Connecting OBIEE11g on Windows to a Kerberos-Secured CDH5 Hadoop Cluster using Cloudera HiveServer2 ODBC Drivers

Rittman Mead Consulting - Sun, 2014-12-28 16:55

In a few previous posts and magazine articles I’ve covered connecting OBIEE11g to a Hadoop cluster, using OBIEE 11.1.1.7 and Cloudera CDH4 and CDH5 as the examples. Things get a bit complicated in that the DataDirect Apache Hive ODBC drivers that Oracle ship are only for HiveServer1 and not the HiveServer2 version that CDH4 and CDH5 use, and the Linux version of OBIEE 11.1.1.7 won’t work with the Cloudera Hive ODBC drivers that you have to use to connect to Hive on CDH4/5. You can however connect OBIEE 11.1.1.7 on Windows to HiveServer2 on CDH4 and CDH5 if you use the Cloudera Hive ODBC drivers for Windows, and although this isn’t supported by Oracle in my experience it does work, albeit with the general OBIEE11g Hive restrictions and caveats detailed in the Metadata Repository Builder’s Guide, and the fact that in-practice Hive is too slow to use for ad-hoc reporting.

However … most enterprise-type customers who run Hadoop on their internal networks have their clusters configured as “secured”, rather than the unsecured cluster examples that you see in most OBIEE connection examples. By default, Hadoop clusters are very trusting of incoming network and client connections and assume that whoever’s connecting is who they say they are, and HDFS and the other cluster components don’t perform any authentication themselves of incoming client connections. In addition, by default all network connections between Hadoop cluster components run in clear text and without any mutual authentication, which is great for a research cluster or PoC but not really appropriate for enterprise customers looking to use Hadoop to store and analyse customer data.

Instead, these customers configure their clusters to run in secured mode, using Kerberos authentication to secure incoming connections, encrypt network traffic and secure connections between the various services in the cluster. How this affects OBIEE though is that your Hive connections through to the cluster also need to use Kerberos authentication, and you (and the OBIEE BI Server) need to have a valid Kerberos ticket when connecting through the Hive ODBC driver. So how do we set this up, and how do we get hold of a secure Hadoop cluster using Kerberos authentication to test against? A few of our customers have asked this question recently, so I thought it’d be worth jotting down a few notes on how to set this up.

At a high-level, if you want to connect OBIEE 11.1.1.7 to a secure, Kerberos-authenticated CDH cluster, there’s three main steps you need to carry out:

  1. Get hold of a Kerberos-secured CDH cluster, and establish the connection details you’ll need to use to connect to it
  2. Make sure the Kerberos server has the correct entries/principals/user details for the user you’re going to securely-connect as
  3. Configure the host environment for OBIEE to work with Kerberos authentication, and then create the connection from OBIEE to the CDH cluster using the correct Kerberos credentials for your user

In my case, I’ve got a Cloudera CDH5.3.0 cluster running in the office that’s been configured to use MIT Kerebos 5 for authentication, set up using an OEL6 VM as the KDC (Key Distribution Centre) and the cluster configured using the new Kerebos setup wizard that was introduced with CDH5.1. Using this wizard automates the creation of the various Kerberos service account and host principals in the Kerberos database, and configures each of the cluster components – YARN, Hive, HDFS and so on – to authenticate with each other using Kerberos authentication and use encrypted network connections for inter-service and inter-node communication.

NewImage

Along with the secured Hadoop cluster, key bits of information and configuration data you’ll need for the OBIEE side are:

  • The krb5.conf file from the Kerberos KDC, which contains details of the Kerberos realm, URL for the KDC server, and other key connection details
  • The name of the Kerberos principal used for the Hive service name on the Hadoop cluster – typically this is “hive”; if you want to connect to Hive first using a JDBC tool such as beeline, you’ll also need the full principal name for this service, in my case “hive/bda3node2.rittmandev.com@RITTMANDEV.COM”
  • The hostname (FQDN) of the node in the CDH cluster that contains the HiveServer2 RPC interface that OBIEE connects to, to run HiveQL queries
  • The Port that HiveServer2 is running on – typically this is “10000”, and the Hive database name (for example, “default’)
  • The name of the Kerebos Realm you’ll be connecting to – for example, MYCOMPANY.COM or in my case, RITTMANDEV.COM (usually in capitals)

In my case, the krb5.conf file that is used to configure Kerebos connections to my KDC looks like this – in your company it might be a bit more complex, but this example defines a simple MIT Kerebos 5 domain:

[logging]
    default = FILE:/var/log/krb5libs.log
    kdc = FILE:/var/log/krb5kdc.log
    admin_server = FILE:/var/log/kadmind.log
[libdefaults]
    default_realm = RITTMANDEV.COM
    dns_lookup_realm = false
    dns_lookup_kdc = false
    ticket_lifetime = 24h
    renew_lifetime = 7d
    forwardable = true
[realms]
    RITTMANDEV.COM = {
    kdc = auth.rittmandev.com
    admin_server = auth.rittmandev.com
}
[domain_realm]
    .rittmandev.com = RITTMANDEV.COM
    rittmandev.com = RITTMANDEV.COM

In my setup, the CDH Hadoop cluster has been configured to use Kerberos authentication for all communications between cluster components and any connections from the outside that use those components; the cluster itself though can still be accessed via unsecured (non-Kerebos authenticated) SSH, though of course this aspect could be secured too. To test out the Hive connectivity before we get into the OBIEE details you can use the beeline CLI that ships with CDH5, and to do this you’ll need to be able to SSH into one of the cluster nodes (if you’ve not got beeline installed on your own workstation) and you’ll need an account (principal) created for you in the Kerebos database to correspond to the Linux user and HDFS/Hive user that has access to the Hive tables you’re interested in. To create such a Kerebos principal for my setup, I used the kadmin.local command on the KDC VM to create a user that matched my Linux/HDFS username and gave it a password:

kadmin.local:  addprinc mrittman
WARNING: no policy specified for mrittman@RITTMANDEV.COM; defaulting to no policy
Enter password for principal "mrittman@RITTMANDEV.COM":
Re-enter password for principal "mrittman@RITTMANDEV.COM":
Principal "mrittman@RITTMANDEV.COM" created.

SSH’ing into one of the secure CDH cluster nodes, I first have to authenticate using the kinit command which when successful, creates a Kerebos ticket that gets cached for a set amount of time, and beeline can thereafter use as part of its own authentication process:

officeimac:.ssh markrittman$ ssh mrittman@bda3node4
mrittman@bda3node4's password: 
[mrittman@bda3node4 ~]$ kinit -p mrittman
Password for mrittman@RITTMANDEV.COM: 
[mrittman@bda3node4 ~]$

Now I can use beeline, and pass the Hive service principal name in the connection details along with the usual host, port and database name. When beeline prompts for my username and password, I use the Kerberos principal name that matches the Linux/HDFS one, and enter that principal’s password:

[mrittman@bda3node4 ~]$ beeline
Beeline version 0.13.1-cdh5.3.0 by Apache Hive
beeline> !connect jdbc:hive2://bda3node2:10000/default;principal=hive/bda3node2.rittmandev.com@RITTMANDEV.COM
scan complete in 2ms
Connecting to jdbc:hive2://bda3node2:10000/default;principal=hive/bda3node2.rittmandev.com@RITTMANDEV.COM
Enter username for jdbc:hive2://bda3node2:10000/default;principal=hive/bda3node2.rittmandev.com@RITTMANDEV.COM: mrittman
Enter password for jdbc:hive2://bda3node2:10000/default;principal=hive/bda3node2.rittmandev.com@RITTMANDEV.COM: ********
Connected to: Apache Hive (version 0.13.1-cdh5.3.0)
Driver: Hive JDBC (version 0.13.1-cdh5.3.0)
Transaction isolation: TRANSACTION_REPEATABLE_READ
0: jdbc:hive2://bda3node2:10000/default> show tables;
+------------------+--+
|     tab_name     |
+------------------+--+
| posts            |
| things_mrittman  |
+------------------+--+
2 rows selected (0.162 seconds)
0: jdbc:hive2://bda3node2:10000/default> select * from things_mrittman;
+---------------------------+-----------------------------+--+
| things_mrittman.thing_id  | things_mrittman.thing_name  |
+---------------------------+-----------------------------+--+
| 1                         | Car                         |
| 2                         | Dog                         |
| 3                         | Hat                         |
+---------------------------+-----------------------------+--+
3 rows selected (0.251 seconds)

So at this point I’ve covered off the first two steps; established the connection details for the secure CDH cluster, and got hold of and confirmed the Kerberos principal details that I’ll need to connect to Hive – now its time to set up the OBIEE element.

In this particular example we’re using Windows to host OBIEE 11.1.1.7, as this is the only platform that we can get the HiveServer2 ODBC drivers to work, in this case the Cloudera Hive ODBC drivers available on their website (free download but registration may be needed). Before we can get this ODBC driver to work though, we need to install the Kerberos client software on the Windows machine so that we can generate the Kerberos ticket that the ODBC driver will need to pass over as part of the authentication process.

To configure the Windows environment for Kerberos authentication, in my case I used the Kerberos for Windows 4.x client software downloadable for free from the MIT website and copied across the krb5.conf file from the KDC server, renaming it to krb5.ini and storing it the default location of c:\ProgramData\MIT\Kerberos5.

NewImage

You also need to define a system environment variable, KRB5CCNAME, to point to a directory where the Kerebos tickets can be cached, in my case I used c:\temp\krb5cache. Once this is done, reboot the Windows environment and you should then be prompted after login to authenticate yourself to the Kerebos KDC.

NewImage

The ticket then stays valid for a set number of days/hours, or you can configure OBIEE itself to authenticate and cache its own ticket – for now though, we’ll create the ticket manually and connect to the secured cluster using these cached ticket details.

After installing the Cloudera Hive ODBC drivers, I create the connection using Kerebos as the Authentication Mechanism, and enter the realm name, HiveServer2 host and the Hive Kerebos principal name, like this:

NewImage

In my case both the BI Administration tool and the OBIEE BI Server were on the same Windows VM, and therefore shared the same ODBC driver install, so I then moved over to the BI Administration tool to import the Hive table metadata details into the RPD and create the physical, logical and presentation layer RPD elements. Depending on how your CDH cluster is set up you might be able to test the connection now by using the View Data… menu item in BI Administration, but in my case I had to do two more things on the CDH cluster itself before I could get Hive queries under this Kerberos principal to run properly.

NewImage

First, as secured CDH Hadoop clusters usually configure HiveServer2 to use “user impersonation” (connecting to Hive as the user you authenticate as, not the user that HiveServer2 authenticates to the Hive service as), YARN and MapReduce jobs run under your account and not the usual “Hive” account that unsecured Hive connections use. Where this causes a problem on CDH installations on RHEL-derived platforms (RHEL, OEL, Centos etc) is that YARN normally blocks jobs running on behalf of users with a UID of <1000 (as this on other Linux distributions typically signifies a system account), RHEL starts user UIDs at 500 and YARN therefore blocks them from running jobs. To fix this, you need to go into Cloudera Manager and edit the YARN configuration settings to lower this UID threshold to something under 500, for example 250:

NewImage

I also needed to alter the group ownership of the temporary directory each node used for the YARN NodeManager’s user files so that YARN could write its temporary files correctly; on each node in the cluster I ran the following Linux commands as root to clear down any files YARN had created before, and recreate the directories with the correct permissions (Hive jobs would fail until I did this, with OBIEE just reporting an ODBC error):

rm -rf /yarn
mkdir -p /yarn/nm
chown -R yarn /yarn
chgrp -R yarn /yarn

Once this is done, queries from the BI Administration tool and from the OBIEE BI Server should connect to the Kerberos-secured CDH cluster successfully, using the Kerberos ticket you obtained using the MIT Kerberos Ticket Manager on login and then passing across the user details under which the YARN, and then Hive job should run.

NewImage

If you’re interested, you can go back to the MIT Kerberos Ticket Manager and see the other Kerberos tickets that were requested and then cached by the Cloudera Hive ODBC driver when it mutually authenticated with the HiveServer2 RPC interface – Kerebos authenticates both ways to ensure that who you’re connecting to is actually who they say they are, in this case checking the HiveServer2 connection you’re connecting to isn’t being spoofed by someone else.

NewImage

So that’s the process for connecting OBIEE to a Kerberos-secured CDH Hadoop cluster in a nutshell; in the New Year I’ll put something together on using Apache Sentry to provide role-based access control for Hive and Impala tables and as of CDH 5.3, HDFS directories, and I’ll also take a look at the new extended ACLs feature in CDH5.2 that goes beyond HDFS’s standard POSIX security model.

Categories: BI & Warehousing

Happy Christmas from Rittman Mead, and Where To Find Us in 2015

Rittman Mead Consulting - Tue, 2014-12-23 09:35

It’s the day before Christmas Eve over in the UK, and we’re finishing up for the Christmas break and looking forward to seeing friends and family over the next few days. To all of our readers, customers, friends and partners, have a great Holiday season and New Year, and we’re looking forward to seeing you at various events and on customer projects in 2015.

If you’re thinking of submitting an abstract for the Rittman Mead BI Forum 2015 the call for papers is now open. with abstracts accepted through to January 18th, 2015. In addition, as well as the BI Forum in May you can also catch-up with us at these events in the first-half of the New Year:

That’s it for now though – have a great Christmas and New Year, and see you in 2015!

Categories: BI & Warehousing

TIMESTAMPS and Presentation Variables

Rittman Mead Consulting - Mon, 2014-12-22 07:38

TIMESTAMPS and Presentation Variables can be some of the most useful tools a report creator can use to invent robust, repeatable reports while maximizing user flexibility.  I intend to transform you into an expert with these functions and by the end of this page you will certainly be able to impress your peers and managers, you may even impress Angus MacGyver.  In this example we will create a report that displays a year over year analysis for any rolling number of periods, by week or month, from any date in time, all determined by the user.  This entire document will only use values from a date and revenue field.

Final Month DS

The TIMESTAMP is an invaluable function that allows a user to define report limits based on a moving target. If the goal of your report is to display Month-to-Date, Year-to-Date, rolling month or truly any non-static period in time, the TIMESTAMP function will allow you to get there.  Often users want to know what a report looked like at some previous point in time, to provide that level of flexibility TIMESTAMPS can be used in conjunction with Presentation Variables.

To create robust TIMESTAMP functions you will first need to understand how the TIMESTAMP works. Take the following example:

Filter Day -7 DS

Here we are saying we want to include all dates greater than or equal to 7 days ago, or from the current date.

  • The first argument, SQL_TSI_DAY, defines the TimeStamp Interval (TSI). This means that we will be working with days.
  • The second argument determines how many of that interval we will be moving, in this case -7 days.
  • The third argument defines the starting point in time, in this example, the current date.

So in the end we have created a functional filter making Date >= 1 week ago, using a TIMESTAMP that subtracts 7 days from today.

Results -7 Days DS

Note: it is always a good practice to include a second filter giving an upper limit like “Time”.”Date” < CURRENT_DATE. Depending on the data that you are working with you might bring in items you don’t want or put unnecessary strain on the system.

We will now start to build this basic filter into something much more robust and flexible.

To start, when we subtracted 7 days in the filter above, let’s imagine that the goal of the filter was to always include dates >= the first of the month. In this scenario, we can use the DAYOFMONTH() function. This function will return the calendar day of any date. This is useful because we can subtract this amount to give us the first of the month from any date by simply subtracting it from that date and adding 1.

Our new filter would look like this:

DayofMonth DS

For example if today is December 18th, DAYOFMONTH(CURRENT_DATE) would equal 18. Thus, we would subtract 18 days from CURRENT_DATE, which is December 18th, and add 1, giving us December 1st.

MTD Dates DS

(For a list of other similar functions like DAYOFYEAR, WEEKOFYEAR etc. click here.)

To make this even better, instead of using CURRENT_DATE you could use a prompted value with the use of a Presentation Variable (for more on Presentation Variables, click here). If we call this presentation variable pDate, for prompted date, our filter now looks like this:

pDate DS

A best practice is to use default values with your presentation variables so you can run the queries you are working on from within your analysis. To add a default value all you do is add the value within braces at the end of your variable. We will use CURRENT_DATE as our default, @{pDate}{CURRENT_DATE}.  Will will refer to this filter later as Filter 1.

{Filter 1}:

pDateCurrentDate DS

As you can see, the filter is starting to take shape. Now lets say we are going to always be looking at a date range of the most recent completed 6 months. All we would need to do is create a nested TIMESTAMP function. To do this, we will “wrap” our current TIMESTAMP with another that will subtract 6 months. It will look like this:

Month -6 DS

Now we have a filter that is greater than or equal to the first day of the month of any given date (default of today) 6 months ago.

Month -6 Result DS

To take this one step further, you can even allow the users to determine the amount of months to include in this analysis by making the value of 6 a presentation variable, we will call it “n” with a default of 6, @{n}{6}.  We will refer to the following filter as Filter 2:

{Filter 2}:

n DS

For more on how to create a prompt with a range of values by altering a current column, like we want to do to allow users to select a value for n, click here.

Our TIMESTAMP function is now fairly robust and will give us any date greater than or equal to the first day of the month from n months ago from any given date. Now we will see what we just created in action by creating date ranges to allow for a Year over Year analysis for any number of months.

Consider the following filter set:

 Robust1 DS

This appears to be pretty intimidating but if we break it into parts we can start to understand its purpose.

Notice we are using the exact same filters from before (Filter 1 and Filter 2).  What we have done here is filtered on two time periods, separated by the OR statement.

The first date range defines the period as being the most recent complete n months from any given prompted date value, using a presentation variable with a default of today, which we created above.

The second time period, after the OR statement, is the exact same as the first only it has been wrapped in another TIMESTAMP function subtracting 1 year, giving you the exact same time frame for the year prior.

YoY Result DS

This allows us to create a report that can run a year over year analysis for a rolling n month time frame determined by the user.

A note on nested TIMESTAMPS:

You will always want to create nested TIMESTAMPS with the smallest interval first. Due to syntax, this will always be the furthest to the right. Then you will wrap intervals as necessary. In this case our smallest increment is day, wrapped by month, wrapped by year.

Now we will start with some more advanced tricks:

  • Instead of using CURRENT_DATE as your default value, use yesterday since most data are only as current as yesterday.  If you use real time or near real time reporting, using CURRENT_DATE may be how you want to proceed. Using yesterday will be valuable especially when pulling reports on the first day of the month or year, you generally want the entire previous time period rather than the empty beginning of a new one.  So, to implement, wherever you have @{pDate}{CURRENT_DATE} replace it with @{pDate}{TIMESTAMPADD(SQL_TSI_DAY,-1,CURRENT_DATE)}
  • Presentation Variables can also be used to determine if you want to display year over year values by month or by week by inserting a variable into your SQL_TSI_MONTH and DAYOFMONTH statements.  Changing MONTH to a presentation variable, SQL_TSI_@{INT}{MONTH} and DAYOF@{INT}{MONTH}, where INT is the name of our variable.  This will require you to create a dummy variable in your prompt to allow users to select either MONTH or WEEK.  You can try something like this: CASE MOD(DAY(“Time”.”Date”),2) WHEN 0 ‘WEEK’ WHEN 1 THEN ‘MONTH’ END

INT DS

MOD DS

DropDown DS

In order for our interaction between Month and Week to run smoothly we have to make one more consideration.  If we are to take the date December 1st, 2014 and subtract one year we get December 1st, 2013, however, if we take the first day of this week, Sunday December 14, 2014 and subtract one year we get Saturday December 14, 2014.  In our analysis this will cause an extra partial week to show up for prior years.  To get around this we will add a case statement determining if ‘@{INT}{MONTH}’ = ‘Week’ THEN subtract 52 weeks from the first of the week ELSE subtract 1 year from the first of the month.

Our final filter set will look like this:

Final Filter DS

With the use of these filters and some creative dashboarding you can end up with a report that easily allows you to view a year over year analysis from any date in time for any number of periods either by month or by week.

Final Month Chart DS

Final Week Chart DS

That really got out of hand in a hurry! Surely, this will impress someone at your work, or even Angus MacGyver, if for nothing less than he or she won’t understand it, but hopefully, now you do!

Also, a colleague of mine Spencer McGhin just wrote a similar article on year over year analyses using a different approach. Feel free to review and consider your options.

 

Calendar Date/Time Functions

These are functions you can use within OBIEE and within TIMESTAMPS to extract the information you need.

  • Current_Date
  • Current_Time
  • Current_TimeStamp
  • Day_Of_Quarter
  • DayName
  • DayOfMonth
  • DayOfWeek
  • DayOfYear
  • Hour
  • Minute
  • Month
  • Month_Of_Quarter
  • MonthName
  • Now
  • Quarter_Of_Year
  • Second
  • TimestampAdd
  • TimestampDiff
  • Week_Of_Quarter
  • Week_Of_Year
  • Year

Back to section

 

Presentation Variables

The only way you can create variables within the presentation side of OBIEE is with the use of presentation variables. They can only be defined by a report prompt. Any value selected by the prompt will then be sent to any references of that filter throughout the dashboard page.

In the prompt:

Pres Var DS

From the “Set a variable” dropdown, select “Presentation Variable”. In the textbox below the dropdown, name your variable (named “n” above).

When calling this variable in your report, use the syntax @{n}{default}

If your variable is a string make sure to surround the variable in single quotes: ‘@{CustomerName]{default}’

Also, when using your variable in your report, it is good practice to assign a default value so that you can work with your report before publishing it to a dashboard. For variable n, if we want a default of 6 it would look like this @{n}{6}

Presentation variables can be called in filters, formulas and even text boxes.

Back to section

 

Dummy Column Prompt

For situations where you would like users to select a numerical value for a presentation variable, like we do with @{n}{6} above, you can convert something like a date field into values up to 365 by using the function DAYOFYEAR(“Time”.”Date”).

As you can see we are returning the SQL Choice List Values of DAYOFYEAR(“Time”.”Date”) <= 52.  Make sure to include an ORDER BY statement to ensure your values are well sorted.

Dummy Script DS

Back to Section

Categories: BI & Warehousing

OBIEE Enterprise Security

Rittman Mead Consulting - Fri, 2014-12-19 05:35

The Rittman Mead Global Services team have recently been involved in a number of security architecture implementations and produced a security model which meets a diverse set of requirements.  Using our experience and standards we have been able to deliver a robust model that addresses the common questions we routinely receive around security, such as :

“Whats considerations do I need to make when exposing Oracle BI to the outside world?”

or

“How can I make a flexible security model which is robust enough to meet the demands of my organisation but easy to maintain?”

The first question is based on a standard enterprise security model where the Oracle BI server is exposed by a web host, enabling SSL and tightening up access security.  This request can be complex to achieve but is something that we have implemented many times now.

The second question is much harder to answer, but our experience has led us to develop a multi-dimensional inheritance security model, with numerous clients that has yielded excellent results.

What is a Multi-dimensional Inheritance Security Model?

The wordy title is actually a simple concept that incorporates 5 key areas:

  • Easy to setup and maintain
  • Flexible
  • Durable
  • Expandable
  • Be consistent throughout the product

While there numerous ways of implementing a security model in Oracle BI, by sticking to the key concepts above, we ensure we get it right.  The largest challenge we face in BI is the different types of security required, and all three need to work in harmony:

  • Application security
  • Content security
  • Data security
Understanding the organisation makeup

The first approach is to consider the makeup of a common organisation and build our security around it.

1

This diagram shows different Departments (Finance, Marketing, Sales) whose data is specific to them, so normally the departmental users should only see their own data that is relevant to them.  In contrast the IT department who are developing the system need visibility across all data and so do the Directors.

 

What types of users do I have?

Next is to consider the types of users we have:

  1. BI Consumer: This will be the most basic and common user who needs to access the system for information.
  2. BI Analyst: As an Analyst the user will be expected to generate more bespoke queries and need ways to represent them. They will also need an area to save these reports.
  3. BI Author: The BI Author will be able to create content and publish that content for the BI Consumers and BI Analysts.
  4. BI Department Admin: The BI Department Admin will be responsible for permissions for their department as well as act as a focal point user.
  5. BI Developer: The BI Developer can be thought of as the person(s) who creates models in the RPD and will need additional access to the system for testing of their models. They might also be responsible for delivering Answers Requests or Dashboards in order to ‘Prove’ the model they created.
  6. BI Administrator:  The Administrator will be responsible for the running of the BI system and will have access to every role.  Most Administrator Task will not require Skills in SQL/Data Warehouse and is generally separated from the BI Developer role.

The types of users here are a combination of every requirement we have seen and might not be required by every client.  The order they are in shows the implied inheritance, so the BI Analyst inherits permissions and privileges from the BI Consumer and so on.

What Types do I need?

Depending on the size of organization determines what types of user groups are required. By default Oracle ships with:

  1. BI Consumer
  2. BI Author
  3. BI Administrator

Typically we would recommend inserting the BI Analyst into the default groups:

  1. BI Consumer
  2. BI Analyst
  3. BI Author
  4. BI Administrator

This works well when there is a central BI team who develop content for the whole organization. The structure would look like this:

2

 

For larger organizations where dashboard development and permissions is handled across multiple BI teams then the BI Department Administrator group can be used to locally manage permissions for each department.  Typically we see the BI team as a central Data Warehouse team who deliver the BI model (RPD) to the multiple BI teams.  In a large Organization the administration of Oracle BI should be handled by someone who isn’t the BI Developer, the structure could look like:

3

 

 

Permissions on groups

Each of the groups will require different permissions, at a high level the permissions would be:

 

Name Permissions BI Consumer
  • View Dashboards
  • Save User Selections
  • Subscribe to Ibots
BI Analyst
  • Access to Answers and standard set of views
  • Some form of storage
  • Access to Subject areas
BI Author
  • Access to Create/Modify Dashboards
  • Save Predefined Sections
  • Access to Action Links
  • Access to Dashboard Prompts
  • Access to BI Publisher
BI Department Admin
  • Ability to apply permissions and manage the Web Catalog
BI Developer
  • Advance access to answers
  • Access to all departments
BI Administrator
  • Everything

 

Understanding the basic security mechanics in 10g and 11g

In Oracle BI 10g the majority of the security is handled in the Oracle BI Server.  This would normally be done through initialisation blocks, which would authenticate the user from a LDAP server, then run a query against a database tables to populate the user into ‘Groups’ used in the RPD and ‘Web Groups’ used in the presentation server.  These groups would have to match in each level; Database, Oracle BI Server and Oracle BI Presentation Server.

With the addition of Enterprise Manager and Weblogic the security elements in Oracle BI 11g radically changed.  Authenticating the user is in the Oracle BI server is no longer the recommended way and is limited in Linux. While the RPD Groups and Presentation Server Web Groups still exist they don’t need to be used.  Users are now authenticated against Weblogic.  This can be done by using Weblogic’s own users and groups or by plugging it into a choice of LDAP servers.  The end result will be Groups and Users that exist in Weblogic.  The groups then need to be mapped to Application Roles in Enterprise Manager, which can be seen by the Oracle BI Presentation Services and Oracle BI Server.  It is recommended to create a one to one mapping for each group.

4

 

What does all this look like then?

Assuming this is for an SME size organization where the Dashboard development (BI Author) is done by the central BI team the groups would like:

 

5

 

The key points are:

  • The generic BI Consumer/Analyst groups give their permissions to the department versions
  • No users should be in the generic BI Consumer/Analyst groups
  • Only users from the BI team should be in the generic BI Author/Administrator group
  • New departments can be easily added
  • the lines denote the inheritance of permissions and privileges

 

Whats next – The Web Catalog?

The setup of the web catalog is very important to ensure that it does not get unwieldy, so it needs to reflect the security model and we would recommend setting up some base folders which look like:

6

 

Each department has their own folder and 4 sub folders. The permissions applied to each department’s root folder is BI Administrators so full control is possible across the top.  This is also true for every folder below however they will have additional explicit permissions described to ensure that the department cannot create any more than the four sub folders.

  • The Dashboard folder is where the dashboards go and the departments BI Developers group will have Full control and the departments BI consumer will have read . This will allow the departments BI Developers to create dashboards,  the departments BI Administrators to apply permissions and the departments consumers and analysts the ability to view.
  • The same permissions are applied to the Dashboard Answers folder to the same effect.
  • The Development Answers folder has Full control given to the departments BI Developers and no access to for the departments BI Analysts or BI Consumers. This folder is mainly for the departments BI Developers to store answers when in the process of development.
  • The Analyst folder is where the departments BI Analysts can save Answers. Therefore they will need full control of this folder.

I hope this article gives some insight into Security with Oracle BI.  Remember that our Global Services products offer a flexible support model where you can harness our knowledge to deliver your projects in a cost effective manner.

Categories: BI & Warehousing

OBIEE and ODI on Hadoop : Next-Generation Initiatives To Improve Hive Performance

Rittman Mead Consulting - Thu, 2014-12-18 16:26

The other week I posted a three-part series (part 1, part 2 and part 3) on going beyond MapReduce for Hadoop-based ETL, where I looked at a typical Apache Pig dataflow-style ETL process and showed how Apache Tez and Apache Spark can potentially make these processes run faster and make better use of in-memory processing. I picked Pig as a data processing environment as the multi-step data transformations creates translate into lots of separate MapReduce jobs in traditional Hadoop ETL environments, but run as a single DAG (directed acyclic graph) under Tez and Spark and can potentially use memory to pass intermediate results between steps, rather than writing all those intermediate datasets to disk.

But tools such as OBIEE and ODI use Apache Hive to interface with the Hadoop world, not Pig, so its improvements to Hive that will have the biggest immediate impact on the tools we use today. And what’s interesting is the developments and work thats going on around Hive in this area, with four different “next-generation Hive” initiatives going on that could end-up making OBIEE and ODI on Hadoop run faster:

  • Hive-on-Tez (or “Stinger”), principally championed by Hortonworks, along with Stinger.next which will enable ACID transactions in HiveQL
  • Hive-on-Spark, a more limited port of Hive to run on Spark and backed by Cloudera amongst others
  • Spark SQL within Apache Spark, which enables SQL queries against Spark RDDs (and Hive tables), and exposes a HiveServer2-compatible Thrift Server for JDBC access
  • Vendor initiatives that build on Hive but are mainly around integration with their RDBMS engines, for example Oracle Big Data SQL

Vendor initiatives like Oracle’s Big Data SQL and Cloudera Impala have the benefit of working now (and are supported), but usually come with some sort of penalty for not working directly within the Hive framework. Oracle’s Big Data SQL, for example, can read data from Hive (very efficiently, using Exadata SmartScan-type technology) but then can’t write-back to Hive, and currently pulls all the Hive data into Oracle if you try and join Oracle and Hive data together. Cloudera’s Impala, on the other hand, is lightening-fast and works directly on the Hadoop platform, but doesn’t support the same ecosystem of SerDes and storage handlers that Hive supports, taking away one of the key flexibility benefits of working with Hive. 

So what about the attempts to extend and improve Hive, or include Hive interfaces and compatibility in Spark? In most cases an ETL routine written as a series of Hive statements isn’t going to be as fast or resource-efficient as a custom Spark program, but if we can make Hive run faster or have a Spark application masquerade as a Hive database, we could effectively give OBIEE and ODI on Hadoop a “free” platform performance upgrade without having to change the way they access Hadoop data. So what are these initiatives about, and how usable are they now with OBIEE and ODI?

Probably the most ambitious next-generation Hive project is the Stinger initiative. Backed by Hortonworks and based on the Apache Tez framework that runs on Hadoop 2.0 and YARN. Stinger aimed first to port Hive to run on Tez (which runs MapReduce jobs but enables them to potentially run as a single DAG), and then add ACID transaction capabilities so that you can UPDATE and DELETE from a Hive table as well as INSERT and SELECT, using a transaction model that allows you to roll-back uncommitted changes (diagram from the Hortonworks Stinger.next page)

NewImage

Tez is more of a set of developer APIs rather than the full data discovery / data analysis platform that Spark aims to provide, but it’s a technology that’s available now as part of Hortonworks HDP2.2 platform and as I showed in the blog post a few days ago, an existing Pig script that you run as-is on a Tez environment typically runs twice as fast as when its using MapReduce to move data around (with usual testing caveats applying, YMMV etc). Hive should be the same as well, giving us the ability to take Hive transformation scripts and run them unchanged except for specifying Tez at the start as the execution engine.

NewImage

Hive on Tez is probably the first of these initiatives we’ll see working with ODI and OBIEE, as ODI has just been certified for Hortonworks HDP2.1, and the new HDP2.2 release is the one that comes with Tez as an option for Pig and Hive query execution. I’m guessing ODI will need to have its Hive KMs updated to add a new option to select Tez or MapReduce as the underlying Hive execution engine, but otherwise I can see this working “out of the box” once ODI support for HDP2.2 is announced.

Going back to the last of the three blog posts I wrote on going beyond MapReduce, many in the Hadoop industry back Spark as the successor to MapReduce rather than Tez as its a more mature implementation that goes beyond the developer-level APIs that Tez aims to provide to make Pig and Hive scripts run faster. As we’ll see in a moment Spark comes with its own SQL capabilities and a Hive-compatible JDBC interface, but the other “swap-out-the-execution-engine” initiative to improve Hive is Hive on Spark, a port of Hive that allows Spark to be used as Hive’s execution engine instead of just MapReduce.

Hive on Spark is at an earlier stage in development than Hive on Tez with the first demo being given at the recent Strata + Hadoop World New York, and specific builds of Spark and versions of Hive needed to get it running. Interestingly though, a post went on the Cloudera Blog a couple of days ago announcing an Amazon AWS AMI machine image that you could use to test Hive on Spark, which though it doesn’t come with a full CDH or HDP installation or features such as a HiveServer JDBC interface, comes with a small TPC-DS dataset and some sample queries that we can use to get a feeling for how it works. I used the AMI image to create an Amazon AWS m3.large instance and gave it a go.

By default, Hive in this demo environment is configured to use Spark as the underlying execution engine. Running a couple of the TPC-DS queries first using this Spark engine, and then switching back to MapReduce by running the command “set hive.execution.engine=mr” within the Hive CLI, I generally found queries using Spark as the execution engine ran 2-3x faster than the MapReduce ones.

NewImage

You can’t read too much into this timing as the demo AMI is really only to show off the functional features (Hive using Spark as the execution engine) and no work on performance optimisation has been done, but it’s encouraging even at this point that it’s significantly faster than the MapReduce version.

Long-term the objective is to have both Tez and Spark available as options as execution engines under Hive, along with MapReduce, as the diagram below from a presentation by Cloudera’s Szenon Ho shows; the advantage of building on Hive like this rather than creating your own new SQL-on-Hadoop engine is that you can make use of the library of SerDes, storage handlers and so on that you’d otherwise need to recreate for any new tool.

NewImage

The third major SQL-on-Hadoop initiative I’ve been looking at is Spark SQL within Apache Spark. Unlike Hive on Spark which aims to swap-out the compiler and execution engine parts of Hive but otherwise leave the rest of the product unchanged, Apache Spark as a whole is a much more freeform, flexible data query and analysis environment that’s aimed more at analysts that business users looking to query their dataset using SQL. That said, Spark has some cool SQL and Hive integration features that make it an interesting platform for doing data analysis and ETL.

In my Spark ETL example the other day, I loaded log data and some reference data into RDDs and then filtered and transformed them using a mix of Scala functions and Spark SQL queries. Running on top of the set of core Spark APIs, Spark SQL allows you to register temporary tables within Spark that map onto RDDs, and give you the option of querying your data using either familiar SQL relational operators, or the more functional programming-style Scala language

NewImage

You can also create connections to the Hive metastore though, and create Hive tables within your Spark application for when you want to persist results to a table rather than work with the temporary tables that Spark SQL usually creates against RDDs. In the code example below, I create a HiveContext as opposed to the sqlContext that I used in the example on the previous blog, and then use that to create a table in my Hive database, running on a Hortonworks HDP2.1 VM with Spark 1.0.0 pre-built for Hadoop 2.4.0:

scala> val hiveContext = new org.apache.spark.sql.hive.HiveContext(sc)
scala> hiveContext.hql("CREATE TABLE posts_hive (post_id int, title string, postdate string, post_type string, author string, post_name string, generated_url string) row format delimited fields terminated by '|' stored as textfile")
scala> hiveContext.hql("LOAD DATA INPATH '/user/root/posts.psv' INTO TABLE posts_hive")

If I then go into the Hive CLI, I can see this new table listed there alongside the other ones:

hive> show tables;
OK
dummy
posts
posts2
posts_hive
sample_07
sample_08
src
testtable2
Time taken: 0.536 seconds, Fetched: 8 row(s)

What’s even more interesting is that Spark also comes with a HiveServer2-compatible Thrift Server, making it possible for tools such as ODI that connect to Hive via JDBC to run Hive queries through Spark, with the Hive metastore providing the metadata but Spark running as the execution engine.

NewImage

This is subtly different to Hive-on-Spark as Hive’s metastore, support for SerDes and storage handlers runs under the covers but Spark provides you with a full programmatic environment, making it possible to just expose Hive tables through the Spark layer, or mix and match data from RDDs, Hive tables and other sources before storing and then exposing the results through the Hive SQL interface. For example then, you could use Oracle SQL*Developer 4.1 with the Cloudera Hive JDBC drivers to connect to this Spark SQL Thrift Server and query the tables just like any other Hive source, but crucially the Hive execution is being done by Spark, rather than MapReduce as would normally happen.

NewImage

Like Hive-on-Spark, Spark SQL and Hive support within Spark SQL are at early stages, with Spark SQL not yet being supported by Cloudera whereas the core Spark API is. From the work I’ve done with it, it’s not yet possible to expose Spark SQL temporary tables through the HiveServer2 Thrift Server interface, and I can’t see a way of creating Hive tables out of RDDs unless you stage the RDD data to a file in-between. But it’s clearly a promising technology and if it becomes possible to seamlessly combine RDD data and Hive data, and expose Spark RDDs registered as tables through the HiveServer2 JDBC interface it could make Spark a very compelling platform for BI and data analyst-type applications. Oracle’s David Allen, for example, blogged about using Spark and the Spark SQL Thrift Server interface to connect ODI to Hive through Spark, and I’d imagine it’d be possible to use the Cloudera HiveServer2 ODBC drivers along with the Windows version of OBIEE 11.1.1.7 to connect to Spark in this way too – if I get some spare time over the Christmas break I’ll try and get an example working.

Categories: BI & Warehousing

Rittman Mead BI Forum 2015 Call for Papers Now Open!

Rittman Mead Consulting - Wed, 2014-12-17 03:00

I’m very pleased to announce that the Call for Papers for the Rittman Mead BI Forum 2015 is now open, with abstract submissions open to January 18th 2015. As in previous years the BI Forum will run over consecutive weeks in Brighton, UK and Atlanta, GA, with the provisional dates and venues as below:

  • Brighton, UK : Hotel Seattle, Brighton, UK : May 6th – 8th 2015
  • Atlanta, GA : Renaissance Atlanta Midtown Hotel, Atlanta, USA : May 13th-15th 2015

Now on it’s seventh year, the Rittman Mead BI Forum is the only conference dedicated entirely to Oracle Business Intelligence, Oracle Business Analytics and the technologies and processes that support it – data warehousing, data analysis, data visualisation, big data and OLAP analysis. We’re looking for session around tips & techniques, project case-studies and success stories, and sessions where you’ve taken Oracle’s BI products and used them in new and innovative ways. Each year we select around eight-to-ten speakers for each event along with keynote speakers and a masterclass session, with speaker choices driven by attendee votes at the end of January, and editorial input from myself, Jon Mead and Charles Elliott and Jordan Meyer.

NewImage

Last year we had a big focus on cloud, and a masterclass and several sessions on bringing Hadoop and big data to the world of OBIEE. This year we’re interested in project stories and experiences around cloud and Hadoop, and we’re keen to hear about any Oracle BI Apps 11g implementations or migrations from the earlier 7.9.x releases. Getting back to basics we’re always interested in sessions around OBIEE, Essbase and data warehouse data modelling, and we’d particularly like to encourage session abstracts on data visualization, BI project methodologies and the incorporation of unstructured, semi-structured and external (public) data sources into your BI dashboards. For an idea of the types of presentations that have been selected in the past, check out the BI Forum 2014, 2013 and 2012 homepages, or feel free to get in touch via email at mark.rittman@rittmanmead.com

The Call for Papers entry form is here, and we’re looking for speakers for Brighton, Atlanta, or both venues if you can speak at both. All session this year will be 45 minutes long, all we’ll be publishing submissions and inviting potential attendees to vote on their favourite sessions towards the end of January. Other than that – have a think about abstract ideas now, and make sure you get them in by January 18th 2015.

Categories: BI & Warehousing

Rittman Mead’s Bar Optimizer

Rittman Mead Consulting - Tue, 2014-12-16 09:08

bar-optimizer-600px

At Rittman Mead R&D, we have the privilege of solving some of our clients’ most challenging data problems. We recently built a set of customized data products that leverage the power of Oracle and Cloudera platforms and wanted to share some of the fun we’ve had in creating unique user experiences. We’ve been thinking about how we can lean on our efforts to help make the holidays even more special for the extended Rittman Mead family. With that inspiration, we had several questions on our minds:

  • How can we throw an amazing holiday party?
  • What gifts can we give that we can be sure our coworkers, friends, and family will enjoy?
  • What gifts would we want for ourselves?

After a discussion over drinks, the answers became clear. We decided to create a tool that uses data analytics to help you create exceptional cocktails for the holidays.

Here is how we did it. First, we analyzed the cocktail recipes of three world-renowned cocktail bars: PDT, Employees Only, and Death & Co. We then turned their drink recipes into data and got to work on the Bar Optimizer, which uses analytics on top of that data to help you make the holiday season tastier than ever before.

To use the Bar Optimizer, enter the liquors and other ingredients that you have on hand to see what drinks you can make. It then recommends additional ingredients that let you create the largest variety of new drinks. You can also use this feature to give great gifts based on others’ liquor cabinets. Finally, try using one of our optimized starter kits to stock your bar for a big holiday party. We’ve crunched the numbers to find the fewest bottles that can make the largest variety of cocktails.

Click the annotated screenshot above for details, and contact us if you would like more information about how we build products that take your data beyond dashboards.

Categories: BI & Warehousing

Linux cluster sysadmin — OS metric monitoring with colmux

Rittman Mead Consulting - Mon, 2014-12-15 12:44

In this mini-series of blog posts I’m taking a look at a few very useful tools that can make your life as the sysadmin of a cluster of Linux machines. This may be a Hadoop cluster, or just a plain simple set of ‘normal’ machines on which you want to run the same commands and monitoring.

First we looked at using SSH keys for intra-machine authorisation, which is a pre-requisite executing the same command across multiple machines using PDSH, as well as what we look at in this article – monitoring OS metrics across a cluster with colmux.

Colmux is written by Mark Seger, the same person who wrote collectl. It makes use of collectl on each target machine to report back OS metrics across a cluster to a single node.

colmux03

Install collectl across the cluster

Using pdsh we can easily install collectl on each node (if it’s not already), which is a pre-requisite for colmux:

pdsh -w root@rnmcluster02-node0[1-4] "yum install -y collectl && service collectl start && chkconfig collectl on"

NB by enabling the collectl service on each node it will capture performance data to file locally, which colmux can replay centrally.

Then install colmux itself, which you can download from Sourceforge. It only needs to be actually installed on a single host, but obviously we could push it out across the cluster with pdsh if we wanted to be able to invoke it on any node at will. Note that here I’m running it on a separate linux box (outside of the cluster) rather than on my Mac:

cd /tmp
# Make sure you get the latest version of collectl-utils, from https://sourceforge.net/projects/collectl-utils/files/
# This example is hardcoded to a version and particular sourceforge mirror
curl -O http://garr.dl.sourceforge.net/project/collectl-utils/collectl-utils-4.8.2/collectl-utils-4.8.2.src.tar.gz
tar xf collectl-utils-4.8.2.src.tar.gz
cd collectl-utils-4.8.2
sudo ./INSTALL
# collectl-utils also includes colplot, so if you might want to use it restart
# apache (assuming it's installed)
sudo service httpd restart

Colmux and networking

Couple of important notes:

  1. The machine you run colmux from needs to have port 2655 open in order for each node’s collectl to send back the data to it.

  2. You also may encounter an issue if you have any odd networking (eg NAT on virtual machines) that causes colmux to not work because it picks the ‘wrong’ network interface of the host to tell collectl on each node to send its data to. Details and workaround here.

colmux in action – real-time view, point in time

Command

colmux -addr 'rnmcluster02-node0[1-4]' -username root

Output

# Mon Dec  1 22:20:40 2014  Connected: 4 of 4
#                    <--------CPU--------><----------Disks-----------><----------Network---------->
#Host                cpu sys inter  ctxsw KBRead  Reads KBWrit Writes   KBIn  PktIn  KBOut  PktOut
rnmcluster02-node01    1   1    28     36      0      0      0      0      0      2      0       2
rnmcluster02-node04    0   0    33     28      0      0     36      8      0      1      0       1
rnmcluster02-node03    0   0    15     17      0      0      0      0      0      1      0       1
rnmcluster02-node02    0   0    18     18      0      0      0      0      0      1      0       1

colmux03

Real-time view, persisted

-cols puts the hosts across the top and time as rows. Specify one or more columns from the output without -cols. In this example it is the values for cpu value, along with the disk read/write (columns 1, 5 and 7 of the metrics as seen above):

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -cols 1,5,7

Output

cpu                            KBRead                         KBWrit
 node01 node02 node03 node04 |  node01 node02 node03 node04 |  node01 node02 node03 node04
      0      0      0      0 |       0      0      0      0 |      12     28      0      0
      0      0      0      0 |       0      0      0      0 |      12     28      0      0
      1      0      1      0 |       0      0      0      0 |       0      0      0      0
      0      0      0      0 |       0      0      0      0 |       0      0      0      0
      0      0      0      0 |       0      0      0      0 |       0      0      0      0
      0      0      0      0 |       0      0      0      0 |       0     20      0      0
      0      0      0      0 |       0      0      0      0 |      52      4      0      0
      0      0      0      2 |       0      0      0      0 |       0      0      0      0
      1      0      0      0 |       0      0      0      0 |       0      0      0      0
     15     16     15     15 |       0      4      4      4 |      20     40     32     48
      0      0      1      1 |       0      0      0      0 |       0      0      4      0
      1      0      0      0 |       0      0      0      0 |       0      0      0      0

colmux04

To check the numbers of the columns that you want to reference, run the command with the --test argument:

colmux -addr 'rnmcluster02-node0[1-4]' -user root --test

>>> Headers <<<
#                    <--------CPU--------><----------Disks-----------><----------Network---------->
#Host                cpu sys inter  ctxsw KBRead  Reads KBWrit Writes   KBIn  PktIn  KBOut  PktOut

>>> Column Numbering <<<
 0 #Host   1 cpu     2 sys     3 inter   4 ctxsw   5 KBRead  6 Reads   7 KBWrit
 8 Writes  9 KBIn   10 PktIn  11 KBOut  12 PktOut

And from there you get the numbers of the columns to reference in the -cols argument.

To include the timestamp, use -oT in the -command and offset the column numbers by 1:

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -cols 2,6,8 -command '-oT'

Output

sys                            Reads                          Writes
#Time    node01 node02 node03 node04 |  node01 node02 node03 node04 |  node01 node02 node03 node04
22:24:50      0      0      0      0 |       0      0      0      0 |       0      0      0      0
22:24:51      1      0      0      0 |       0      0      0      0 |       0      0      0      0
22:24:52      0      0      0      0 |       0      0      0      0 |       0     16      0     16
22:24:53      1      0      0      0 |       0      0      0      0 |      36      0     16      0
22:24:54      0      0      0      1 |       0      0      0      0 |       0      0      0      0
22:24:55      0      0      0      0 |       0      0      0      0 |       0     20     32     20

NB There’s a bug with colmux 4.8.2 that prevents you accessing the first metric with -cols when you also enable timestamp -oTdetails here.

Specifying additional metrics

Collectl (which is what colmux calls to get the data) can fetch metrics from multiple subsystems on a node. You can access all of these through colmux too. By default when you run colmux you get cpu, disk and network but you can specify others using the -s argument followed by the subsystem identifier.

To examine the available subsystems run collectl on one of the target nodes:

[root@rnmcluster02-node01 ~]# collectl --showsubsys
The following subsystems can be specified in any combinations with -s or
--subsys in both record and playbackmode.  [default=bcdfijmnstx]

These generate summary, which is the total of ALL data for a particular type
  b - buddy info (memory fragmentation)
  c - cpu
  d - disk
  f - nfs
  i - inodes
  j - interrupts by CPU
  l - lustre
  m - memory
  n - network
  s - sockets
  t - tcp
  x - interconnect (currently supported: OFED/Infiniband)
  y - slabs

From the above list we can see that if we want to also show memory detail alongside CPU we need to include m and c in the subsystem list:

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -command '-scm'

Output

# Tue Dec  2 08:02:38 2014  Connected: 4 of 4
#                    <--------CPU--------><-----------Memory----------->
#Host                cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map
rnmcluster02-node02    1   0    19     18  33M  15M 345M 167M  30M  56M
rnmcluster02-node04    0   0    30     24  32M  15M 345M 167M  30M  56M
rnmcluster02-node03    0   0    30     36  32M  15M 345M 165M  30M  56M
rnmcluster02-node01    0   0    16     16  29M  15M 326M 167M  27M  81M

Changing the sample frequency

To change the sample frequency use the -i syntax in -command:

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -command '-scm -i10 -oT' -cols 2,4

Samples every 10 seconds:

sys                            ctxsw
#Time    node01 node02 node03 node04 |  node01 node02 node03 node04
08:06:29     -1     -1     -1     -1 |      -1     -1     -1     -1
08:06:39     -1     -1     -1     -1 |      -1     -1     -1     -1
08:06:49      0      0      0      0 |      14     13     15     19
08:06:59      0      0      0      0 |      13     13     17     21
08:07:09      0      0      0      0 |      19     18     15     24
08:07:19      0      0      0      0 |      13     13     15     19
08:07:29      0      0      0      0 |      13     13     14     19
08:07:39      0      0      0      0 |      12     13     13     19

Column width

Add the -colwidth argument

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -command '-scm' -cols 1 -colwidth 20

Output

cpu
  rnmcluster02-node01  rnmcluster02-node02  rnmcluster02-node03  rnmcluster02-node04
                   -1                   -1                   -1                   -1
                   -1                   -1                   -1                   -1
                    1                    0                    0                    0
                    0                    0                    0                    0
                    0                    1                    0                    0
                    0                    0                    1                    0
                    1                    0                    1                    0
                    0                    1                    0                    0

Playback

As well as running interactively, collectl can run as a service and record metric samples to disk. Using colmux you can replay these from across the cluster.

Within the -command, include -p and the path to the collectl log files (assumes that it is the same on each host). As with real-time mode, for different subsystems change the flags after -s

Command

colmux -addr 'rnmcluster02-node0[1-4]' -user root -command '-p /var/log/collectl/*20141201* -scmd -oD'

Output

[...]
# 21:48:50  Reporting: 4 of 4
#                                      <--------CPU--------><-----------Memory-----------><----------Disks----------->
#Host               Date     Time      cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map KBRead  Reads KBWrit Writes
rnmcluster02-node04 20141201 21:48:50    0   0    17     15  58M  10M 340M 162M  30M  39M      0      0      1      0
rnmcluster02-node03 20141201 21:48:50    0   0    11     13  58M  10M 340M 160M  30M  39M      0      0      0      0
rnmcluster02-node02 20141201 21:48:50    0   0    11     15  58M  10M 340M 163M  29M  39M      0      0      1      0
rnmcluster02-node01 20141201 21:48:50    0   0    12     14  33M  12M 342M 157M  27M  63M      0      0      1      0

# 21:49:00  Reporting: 4 of 4
#                                      <--------CPU--------><-----------Memory-----------><----------Disks----------->
#Host               Date     Time      cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map KBRead  Reads KBWrit Writes
rnmcluster02-node04 20141201 21:49:00    0   0    17     15  58M  10M 340M 162M  30M  39M      0      0      4      0
rnmcluster02-node03 20141201 21:49:00    0   0    13     14  58M  10M 340M 160M  30M  39M      0      0      5      0
rnmcluster02-node02 20141201 21:49:00    0   0    12     14  58M  10M 340M 163M  29M  39M      0      0      1      0
rnmcluster02-node01 20141201 21:49:00    0   0    12     15  33M  12M 342M 157M  27M  63M      0      0      6      0

# 21:49:10  Reporting: 4 of 4
#                                      <--------CPU--------><-----------Memory-----------><----------Disks----------->
#Host               Date     Time      cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map KBRead  Reads KBWrit Writes
rnmcluster02-node04 20141201 21:49:10    0   0    23     23  58M  10M 340M 162M  30M  39M      0      0      1      0
rnmcluster02-node03 20141201 21:49:10    0   0    19     24  58M  10M 340M 160M  30M  39M      0      0      2      0
rnmcluster02-node02 20141201 21:49:10    0   0    18     23  58M  10M 340M 163M  29M  39M      0      0      2      1
rnmcluster02-node01 20141201 21:49:10    0   0    18     24  33M  12M 342M 157M  27M  63M      0      0      1      0
[...]

Restrict the time frame by adding to -command the arguments -from and/or -thru

[oracle@rnm-ol6-2 ~]$ colmux -addr 'rnmcluster02-node0[1-4]' -user root -command '-p /var/log/collectl/*20141201* -scmd -oD --from 21:40:00 --thru 21:40:10'
# 21:40:00  Reporting: 4 of 4
#                                      <--------CPU--------><-----------Memory-----------><----------Disks----------->
#Host               Date     Time      cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map KBRead  Reads KBWrit Writes
rnmcluster02-node04 20141201 21:40:00    0   0    16     14  59M  10M 340M 162M  30M  39M      0      0      0      0
rnmcluster02-node03 20141201 21:40:00    0   0    12     14  58M  10M 340M 160M  30M  39M      0      0      8      1
rnmcluster02-node02 20141201 21:40:00    0   0    12     15  59M  10M 340M 162M  30M  39M      0      0      6      1
rnmcluster02-node01 20141201 21:40:00    0   0    13     16  56M  11M 341M 156M  27M  42M      0      0      7      1

# 21:40:10  Reporting: 4 of 4
#                                      <--------CPU--------><-----------Memory-----------><----------Disks----------->
#Host               Date     Time      cpu sys inter  ctxsw Free Buff Cach Inac Slab  Map KBRead  Reads KBWrit Writes
rnmcluster02-node04 20141201 21:40:10    0   0    26     33  59M  10M 340M 162M  30M  39M      1      0     10      2
rnmcluster02-node03 20141201 21:40:10    0   0    20     31  58M  10M 340M 160M  30M  39M      0      0      4      1
rnmcluster02-node02 20141201 21:40:10    0   0    23     35  59M  10M 340M 162M  30M  39M      3      0      9      2
rnmcluster02-node01 20141201 21:40:10    0   0    23     37  56M  11M 341M 156M  27M  42M      4      1      4      1


[oracle@rnm-ol6-2 ~]$

colmux reference

You can find more about colmux from the website:

as well as the built in man page man colmux

Visualising collectl data with colplot

As a little bonus to the above, colmux is part of the collectl-utils package, which also includes colplot, a gnuplot-based web tool that renders collectl data into graphs. It’s pretty easy to set up, running under Apache just fine and just needing gnuplot installed if you haven’t already. It can report metrics across a cluster if you make sure that you first make each node’s collectl data available locally to colplot.

Navigating to the web page shows the interface from which you can trigger graph plots based on the collectl data available:

colplot’s utilitarian graphs are a refreshing contrast to every webapp that is built nowadays promising “beautiful” visualisations (which no doubt the authors are “passionate” about making “awesome”):

The graphs are functional and can be scaled as needed, but each change is a trip back to the front page to tweak options and re-render:

colplot

For me, colplot is an excellent tool for point-in-time analysis and diagnostics, but for more generalised monitoring with drilldown into detail, it is too manual to be viable and I’ll be sticking with collectl -> graphite -> grafana with its interactive and flexible graph rendering:

grafana

Do note however that colplot specifically does not drop data points, so if there is a spike in your data you will see it. Other tools (possibly including graphite but I’ve not validated this) will, for larger timespans average out data series so as to provide a smoother picture of a metric (eg instead of a point every second, maybe every ten seconds). If you are doing close analysis of a system’s behaviour in a particular situation this may be a problem. If you are wanting more generalised overview of a system’s health, with the option to drill into data historical as needed, it will be less of an issue.

Summary

When working with multiple Linux machines I would first and foremost make sure SSH keys are set up in order to ease management through password-less logins.

After SSH keys, I would recommend pdsh for parallel execution of the same SSH command across the cluster. It’s a big time saver particularly when initially setting up the cluster given the installation and configuration changes that are inevitably needed.

To monitor a cluster I would always recommend collectl as the base metric collector. colmux works excellently for viewing these metrics from across the cluster in a single place from the commandline. For viewing the metrics over the longer term you can either store them in (or replay them into) Graphite/Carbon, and render them in Grafana. You have the option of colplot too since this is installed as part of colmux.

So now your turn – what particular tools or tips do you have for working with a cluster of Linux machines? Leave your answers in the comments below, or tweet them to me at @rmoff.

Categories: BI & Warehousing

Paginated HTML is here and has been for some time ... I think!

Tim Dexter - Fri, 2014-12-12 18:03

We have a demo environment in my team and of course things get a little beaten up in there. Our go to, 'here's Publisher' report was looking really bad. Data was not returning or being rendered correctly on the five templates we have for it.
So, I spent about a half hour cleaning up the report; getting things working again; clearing out the rubbish. I noticed that one of the layouts when rendered in HTML was repeatedly showing a header down the screen. Oh, I know where to get rid of that and off I click to the report properties to fix it. But what is this I see? Is it? Can it be? Are my tired old eyes deceiving me?

Yes, Dexter, you see that right, 'View Paginated'! I nervously changed the value to 'true' and went back to the HTML output.
Holy Amaze Balls Batman, paginated HTML, the holy grail of HTML rendered reports, the Mount Everest of ... no, thats too easy, the K2 of html output ... its fan-bloody-tastic! Can you tell Im excited? I was immediately on messenger to Leslie (doc writer extraordinaire) 


Obviously not quite as big a deal in the sane, real world outside of my head. 'Oh yeah, we have that now ...' Leslie is so calm and collected, however, she does like Maroon 5 but, we overlook that :)

I command you 11.1.1.6+'ers to go find the property and turn it on right now and bask in the glory that is, 'paginated html.!'
I cannot stop clicking back and forth and then to the end and then all the way back to the beginning. Its fantastic!

Just look at those icons, just click em, you know you want to!

Categories: BI & Warehousing

File Encoding in the Next Generation Outline Extractor

Tim Tow - Tue, 2014-12-09 20:11
We had a couple of issues reported with the output of the Next Generation Outline Extractor where the exported file did not work properly as a load file. After some investigation, we found that the file encoding was incorrect. We were encoding the files using the Unicode/UTF-8 format. We chose this encoding so that we could write all characters in all languages, but we did not consider that UTF-8 is only valid for loading Unicode databases in Essbase.

To resolve this issue, we decided to add a configuration to the Next Generation Outline Extractor to allow users to select the file encoding. Here is a screenshot showing the new configuration setting.



As of yesterday, December 8, 2014, the updated Next Generation Outline Extractor is available on our website. The first version to feature this functionality is version 2.0.2.692. Version 2.0.2.692 is available for all Essbase versions from Essbase 9.3.1 forward. We are also happy to announce that his version of the Next Generation Outline Extractor is the first version to support the recently released Essbase 11.1.2.3.505.

If you encounter any issues with the Next Generation Outline Extractor, please don't hesitate to contact our support team at support@appliedolap.com.

Categories: BI & Warehousing

Linux cluster sysadmin — Parallel command execution with PDSH

Rittman Mead Consulting - Tue, 2014-12-09 11:40

In this series of blog posts I’m taking a look at a few very useful tools that can make your life as the sysadmin of a cluster of Linux machines easier. This may be a Hadoop cluster, or just a plain simple set of ‘normal’ machines on which you want to run the same commands and monitoring.

Previously we looked at using SSH keys for intra-machine authorisation, which is a pre-requisite what we’ll look at here — executing the same command across multiple machines using PDSH. In the next post of the series we’ll see how we can monitor OS metrics across a cluster with colmux.

PDSH is a very smart little tool that enables you to issue the same command on multiple hosts at once, and see the output. You need to have set up ssh key authentication from the client to host on all of them, so if you followed the steps in the first section of this article you’ll be good to go.

The syntax for using it is nice and simple:

  • -w specifies the addresses. You can use numerical ranges [1-4] and/or comma-separated lists of hosts. If you want to connect as a user other than the current user on the calling machine, you can specify it here (or as a separate -l argument)
  • After that is the command to run.

For example run against a small cluster of four machines that I have:

robin@RNMMBP $ pdsh -w root@rnmcluster02-node0[1-4] date

rnmcluster02-node01: Fri Nov 28 17:26:17 GMT 2014
rnmcluster02-node02: Fri Nov 28 17:26:18 GMT 2014
rnmcluster02-node03: Fri Nov 28 17:26:18 GMT 2014
rnmcluster02-node04: Fri Nov 28 17:26:18 GMT 2014

PDSH can be installed on the Mac under Homebrew (did I mention that Rittman Mead laptops are Macs, so I can do all of this straight from my work machine… :-) )

brew install pdsh

And if you want to run it on Linux from the EPEL yum repository (RHEL-compatible, but packages for other distros are available):

yum install pdsh

You can run it from a cluster node, or from your client machine (assuming your client machine is mac/linux).

Example – install and start collectl on all nodes

I started looking into pdsh when it came to setting up a cluster of machines from scratch. One of the must-have tools I like to have on any machine that I work with is the excellent collectl. This is an OS resource monitoring tool that I initially learnt of through Kevin Closson and Greg Rahn, and provides the kind of information you’d get from top etc – and then some! It can run interactively, log to disk, run as a service – and it also happens to integrate very nicely with graphite, making it a no-brainer choice for any server.

So, instead of logging into each box individually I could instead run this:

pdsh -w root@rnmcluster02-node0[1-4] yum install -y collectl
pdsh -w root@rnmcluster02-node0[1-4] service collectl start
pdsh -w root@rnmcluster02-node0[1-4] chkconfig collectl on

Yes, I know there are tools out there like puppet and chef that are designed for doing this kind of templated build of multiple servers, but the point I want to illustrate here is that pdsh enables you to do ad-hoc changes to a set of servers at once. Sure, once I have my cluster built and want to create an image/template for future builds, then it would be daft if I were building the whole lot through pdsh-distributed yum commands.

Example – setting up the date/timezone/NTPD

Often the accuracy of the clock on each server in a cluster is crucial, and we can easily do this with pdsh:

Install packages

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] yum install -y ntp ntpdate

Set the timezone:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] ln -sf /usr/share/zoneinfo/Europe/London /etc/localtime

Force a time refresh:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] ntpdate pool.ntp.org
rnmcluster02-node03: 30 Nov 20:46:22 ntpdate[27610]: step time server 176.58.109.199 offset -2.928585 sec
rnmcluster02-node02: 30 Nov 20:46:22 ntpdate[28527]: step time server 176.58.109.199 offset -2.946021 sec
rnmcluster02-node04: 30 Nov 20:46:22 ntpdate[27615]: step time server 129.250.35.250 offset -2.915713 sec
rnmcluster02-node01: 30 Nov 20:46:25 ntpdate[29316]: 178.79.160.57 rate limit response from server.
rnmcluster02-node01: 30 Nov 20:46:22 ntpdate[29316]: step time server 176.58.109.199 offset -2.925016 sec

Set NTPD to start automatically at boot:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] chkconfig ntpd on

Start NTPD:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] service ntpd start

Example – using a HEREDOC (here-document) and sending quotation marks in a command with PDSH

Here documents (heredocs) are a nice way to embed multi-line content in a single command, enabling the scripting of a file creation rather than the clumsy instruction to “open an editor and paste the following lines into it and save the file as /foo/bar”.

Fortunately heredocs work just fine with pdsh, so long as you remember to enclose the whole command in quotation marks. And speaking of which, if you need to include quotation marks in your actual command, you need to escape them with a backslash. Here’s an example of both, setting up the configuration file for my ever-favourite gnu screen on all the nodes of the cluster:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] "cat > ~/.screenrc <<EOF
hardstatus alwayslastline \"%{= RY}%H %{kG}%{G} Screen(s): %{c}%w %=%{kG}%c  %D, %M %d %Y  LD:%l\"
startup_message off
msgwait 1
defscrollback 100000
nethack on
EOF
"

Now when I login to each individual node and run screen, I get a nice toolbar at the bottom:

Combining commands

To combine commands together that you send to each host you can use the standard bash operator semicolon ;

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] "date;sleep 5;date"
rnmcluster02-node01: Sun Nov 30 20:57:06 GMT 2014
rnmcluster02-node03: Sun Nov 30 20:57:06 GMT 2014
rnmcluster02-node04: Sun Nov 30 20:57:06 GMT 2014
rnmcluster02-node02: Sun Nov 30 20:57:06 GMT 2014
rnmcluster02-node01: Sun Nov 30 20:57:11 GMT 2014
rnmcluster02-node03: Sun Nov 30 20:57:11 GMT 2014
rnmcluster02-node04: Sun Nov 30 20:57:11 GMT 2014
rnmcluster02-node02: Sun Nov 30 20:57:11 GMT 2014

Note the use of the quotation marks to enclose the entire command string. Without them the bash interpretor will take the ; as the delineator of the local commands, and try to run the subsequent commands locally:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node0[1-4] date;sleep 5;date
rnmcluster02-node03: Sun Nov 30 20:57:53 GMT 2014
rnmcluster02-node04: Sun Nov 30 20:57:53 GMT 2014
rnmcluster02-node02: Sun Nov 30 20:57:53 GMT 2014
rnmcluster02-node01: Sun Nov 30 20:57:53 GMT 2014
Sun 30 Nov 2014 20:58:00 GMT

You can also use && and || to run subsequent commands conditionally if the previous one succeeds or fails respectively:

robin@RNMMBP $ pdsh -w root@rnmcluster02-node[01-4] "chkconfig collectl on && service collectl start"

rnmcluster02-node03: Starting collectl: [  OK  ]
rnmcluster02-node02: Starting collectl: [  OK  ]
rnmcluster02-node04: Starting collectl: [  OK  ]
rnmcluster02-node01: Starting collectl: [  OK  ]

Piping and file redirects

Similar to combining commands above, you can pipe the output of commands, and you need to use quotation marks to enclose the whole command string.

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] "chkconfig|grep collectl"
rnmcluster02-node03: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node01: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node04: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node02: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off

However, you can pipe the output from pdsh to a local process if you want:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] chkconfig|grep collectl
rnmcluster02-node02: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node04: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node03: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off
rnmcluster02-node01: collectl           0:off   1:off   2:on    3:on    4:on    5:on    6:off

The difference is that you’ll be shifting the whole of the pipe across the network in order to process it locally, so if you’re just grepping etc this doesn’t make any sense. For use of utilities held locally and not on the remote server though, this might make sense.

File redirects work the same way – within quotation marks and the redirect will be to a file on the remote server, outside of them it’ll be local:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] "chkconfig>/tmp/pdsh.out"
robin@RNMMBP ~ $ ls -l /tmp/pdsh.out
ls: /tmp/pdsh.out: No such file or directory

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] chkconfig>/tmp/pdsh.out
robin@RNMMBP ~ $ ls -l /tmp/pdsh.out
-rw-r--r--  1 robin  wheel  7608 30 Nov 19:23 /tmp/pdsh.out

Cancelling PDSH operations

As you can see from above, the precise syntax of pdsh calls can be hugely important. If you run a command and it appears ‘stuck’, or if you have that heartstopping realisation that the shutdown -h now you meant to run locally you ran across the cluster, you can press Ctrl-C once to see the status of your commands:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] sleep 30
^Cpdsh@RNMMBP: interrupt (one more within 1 sec to abort)
pdsh@RNMMBP:  (^Z within 1 sec to cancel pending threads)
pdsh@RNMMBP: rnmcluster02-node01: command in progress
pdsh@RNMMBP: rnmcluster02-node02: command in progress
pdsh@RNMMBP: rnmcluster02-node03: command in progress
pdsh@RNMMBP: rnmcluster02-node04: command in progress

and press it twice (or within a second of the first) to cancel:

robin@RNMMBP ~ $ pdsh -w root@rnmcluster02-node[01-4] sleep 30
^Cpdsh@RNMMBP: interrupt (one more within 1 sec to abort)
pdsh@RNMMBP:  (^Z within 1 sec to cancel pending threads)
pdsh@RNMMBP: rnmcluster02-node01: command in progress
pdsh@RNMMBP: rnmcluster02-node02: command in progress
pdsh@RNMMBP: rnmcluster02-node03: command in progress
pdsh@RNMMBP: rnmcluster02-node04: command in progress
^Csending SIGTERM to ssh rnmcluster02-node01
sending signal 15 to rnmcluster02-node01 [ssh] pid 26534
sending SIGTERM to ssh rnmcluster02-node02
sending signal 15 to rnmcluster02-node02 [ssh] pid 26535
sending SIGTERM to ssh rnmcluster02-node03
sending signal 15 to rnmcluster02-node03 [ssh] pid 26533
sending SIGTERM to ssh rnmcluster02-node04
sending signal 15 to rnmcluster02-node04 [ssh] pid 26532
pdsh@RNMMBP: interrupt, aborting.

If you’ve got threads yet to run on the remote hosts, but want to keep running whatever has already started, you can use Ctrl-C, Ctrl-Z:

robin@RNMMBP ~ $ pdsh -f 2 -w root@rnmcluster02-node[01-4] "sleep 5;date"
^Cpdsh@RNMMBP: interrupt (one more within 1 sec to abort)
pdsh@RNMMBP:  (^Z within 1 sec to cancel pending threads)
pdsh@RNMMBP: rnmcluster02-node01: command in progress
pdsh@RNMMBP: rnmcluster02-node02: command in progress
^Zpdsh@RNMMBP: Canceled 2 pending threads.
rnmcluster02-node01: Mon Dec  1 21:46:35 GMT 2014
rnmcluster02-node02: Mon Dec  1 21:46:35 GMT 2014

NB the above example illustrates the use of the -f argument to limit how many threads are run against remote hosts at once. We can see the command is left running on the first two nodes and returns the date, whilst the Ctrl-C – Ctrl-Z stops it from being executed on the remaining nodes.

PDSH_SSH_ARGS_APPEND

By default, when you ssh to new host for the first time you’ll be prompted to validate the remote host’s SSH key fingerprint.

The authenticity of host 'rnmcluster02-node02 (172.28.128.9)' can't be established.
RSA key fingerprint is 00:c0:75:a8:bc:30:cb:8e:b3:8e:e4:29:42:6a:27:1c.
Are you sure you want to continue connecting (yes/no)?

This is one of those prompts that the majority of us just hit enter at and ignore; if that includes you then you will want to make sure that your PDSH call doesn’t fall in a heap because you’re connecting to a bunch of new servers all at once. PDSH is not an interactive tool, so if it requires input from the hosts it’s connecting to it’ll just fail. To avoid this SSH prompt, you can set up the environment variable PDSH_SSH_ARGS_APPEND as follows:

export PDSH_SSH_ARGS_APPEND="-q -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null"

The -q makes failures less verbose, and the -o passes in a couple of options, StrictHostKeyChecking to disable the above check, and UserKnownHostsFile to stop SSH keeping a list of host IP/hostnames and corresponding SSH fingerprints (by pointing it at /dev/null). You’ll want this if you’re working with VMs that are sharing a pool of IPs and get re-used, otherwise you get this scary failure:

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@    WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!     @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!
Someone could be eavesdropping on you right now (man-in-the-middle attack)!
It is also possible that a host key has just been changed.
The fingerprint for the RSA key sent by the remote host is
00:c0:75:a8:bc:30:cb:8e:b3:8e:e4:29:42:6a:27:1c.
Please contact your system administrator.

For both of these above options, make sure you’re aware of the security implications that you’re opening yourself up to. For a sandbox environment I just ignore them; for anything where security is of importance make sure you are aware of quite which server you are connecting to by SSH, and protecting yourself from MitM attacks.

PDSH Reference

You can find out more about PDSH at https://code.google.com/p/pdsh/wiki/UsingPDSH

Summary

When working with multiple Linux machines I would first and foremost make sure SSH keys are set up in order to ease management through password-less logins.

After SSH keys, I would recommend pdsh for parallel execution of the same SSH command across the cluster. It’s a big time saver particularly when initially setting up the cluster given the installation and configuration changes that are inevitably needed.

In the next article of this series we’ll see how the tool colmux is a powerful way to monitor OS metrics across a cluster.

So now your turn – what particular tools or tips do you have for working with a cluster of Linux machines? Leave your answers in the comments below, or tweet them to me at @rmoff.

Categories: BI & Warehousing

Linux cluster sysadmin — SSH keys

Rittman Mead Consulting - Tue, 2014-12-09 05:34

In this short series of blog posts I’m going to take a look at a few very useful tools that can make your life as the sysadmin of a cluster of Linux machines easier. This may be a Hadoop cluster, or just a plain simple set of ‘normal’ machines on which you want to run the same commands and monitoring.

To start with, we’re going to use the ever-awesome ssh keys to manage security on the cluster. After that we’ll look at executing the same command across multiple machines at the same time using PDSH, and then monitoring OS metrics across a cluster with colmux.

In a nutshell, ssh keys enable us to do password-less authentication in a secure way. You can find a detailed explanation of them in a previous post that I wrote, tips and tricks for OBIEE Linux sysadmin. Beyond the obvious time-saving function of not having to enter a password each time we connect to a machine, having SSH keys in place enable the use of the tools we discuss later, pdsh and colmux.

Working with SSH keys involves taking the public key from a pair, and adding that to another machine in order to allow the owner of the pair’s private key to access that machine. What we’re going to do here is generate a unique key pair that will be used as the identity across the cluster. So each node will have a copy of the private key, in order to be able to authenticate to any other node, which will be holding a copy of the public key (as well as, in turn, the same private key).

In this example I’m going to use my own client machine to connect to the cluster. You could easily use any of the cluster nodes too if a local machine would not be appropriate.
As a side-note, this is another reason why I love the fact that Rittman Mead standard-issue laptop is a MacBook, and just under the covers of Mac OS is a *nix-based command-line meaning that a lot of sysadmin work can be done natively without needing additional tools that you would on Windows (e.g. PuTTY, WinSCP, Pageant, etc etc).

SSH key strategy

We’ve several ways we could implement the SSH keys. Because it’s a purely sandbox cluster, I could use the same SSH key pair that I generate for the cluster on my machine too, so the same public/private key pair is distributed thus:

If we wanted a bit more security, a better approach might be to distribute my personal SSH key’s public key across the cluster too, and leave the cluster’s private key to truly identify cluster nodes alone. An additional benefit of this approach is that is the client does not need to hold a copy of the cluster’s SSH private key, instead just continuing to use their own.

For completeness, the extreme version of the key strategy would be for each machine to have its own ssh key pair (i.e. its own security identity), with the corresponding public keys distributed to the other nodes in the cluster:

But anyway, here we’re using the second option – a unique keypair used across the cluster and the client’s public ssh key distributed across the cluster too.

Generating the SSH key pair

First, we need to generate the key. I’m going to create a folder to hold it first, because in a moment we’re going to push it and a couple of other files out to all the servers in the cluster and it’s easiest to do this from a single folder.

mkdir /tmp/rnmcluster02-ssh-keys

Note that in the ssh-keygen command below I’m specifying the target path for the key with the -f argument; if you don’t then watch out that you don’t accidentally overwrite your own key pair in the default path of ~/.ssh.

The -q -N "" flags instruct the key generation to use no passphrase for the key and to not prompt for it either. This is the lowest friction approach (you don’t need to unlock the ssh key with a passphrase before use) but also the least secure. If you’re setting up access to a machine where security matters then bear in mind that without a passphrase on an ssh key anyone who obtains it can therefore access any machine to which the key has been granted access (i.e. on which its public key has been deployed).

ssh-keygen -f /tmp/rnmcluster02-ssh-keys/id_rsa -q -N ""

This generates in the tmp folder two files – the private and public (.pub) keys of the pair:

robin@RNMMBP ~ $ ls -l /tmp/rnmcluster02-ssh-keys
total 16
-rw-------  1 robin  wheel  1675 30 Nov 17:28 id_rsa
-rw-r--r--  1 robin  wheel   400 30 Nov 17:28 id_rsa.pub

Preparing the authorized_keys file

Now we’ll prepare the authorized_keys file which is where the public SSH key of any identity permitted to access the machine is stored. Note that each user on a machine has their own authorized_keys file, in ~/.ssh/. So for example, the root user has the file in /root/.ssh/authorized_keys and any public key listed in that file will be able to connect to the server as the root user. Be aware the American [mis-]spelling of “authorized” – spell it [correctly] as “authorised” and you’ll not get any obvious errors, but the ssh key login won’t work either.

So we’re going to copy the public key of the unique pair that we just created for the cluster into the authorized_keys file. In addition we will copy in our own personal ssh key (and any other public key that we want to give access to all the nodes in the cluster):

cp /tmp/rnmcluster02-ssh-keys/id_rsa.pub /tmp/rnmcluster02-ssh-keys/authorized_keys
# [optional] Now add any other keys (such as your own) into the authorized_keys file just created
cat ~/.ssh/id_rsa.pub >> /tmp/rnmcluster02-ssh-keys/authorized_keys
# NB make sure the previous step is a double >> not > since the double appends to the file, a single overwrites.

Distributing the SSH artefacts

Now we’re going to push this set of SSH files out to the .ssh folder of the target user on each node, which in this case is the root user. From a security point of view it’s probably better to use a non-root user for login and then sudo as required, but we’re keeping things simple (and less secure) to start with here. So the files in our folder are:

  • id_rsa – the private key of the key pair
  • id_rsa.pub – the public key of the key pair. Strictly speaking this doesn’t need distributing to all nodes, but it’s conventional and handy to hold it alongside the private key.
  • authorized_keys – this is the file that the sshd daemon on each node will look at to validate an incoming login request’s offered private key, and so needs to hold the public key of anyone who is allowed to access the machine as this user.

To copy the files we’ll use scp, but how you get them in place doesn’t really matter so much, so long as they get to the right place:

scp -r /tmp/rnmcluster02-ssh-keys root@rnmcluster02-node01:~/.ssh

At this point you’ll need to enter the password for the target user, but rejoice! This is the last time you’ll need to enter it as subsequent logins will be authenticated using the ssh keys that you’re now configuring.

Run the scp for all nodes in the cluster. If you’ve four nodes in the cluster your output should look something like this:

$ scp -r /tmp/rnmcluster02-ssh-keys/ root@rnmcluster02-node01:~/.ssh
root@rnmcluster02-node01's password:
authorized_keys                                                  100%  781     0.8KB/s   00:00
id_rsa                                                           100% 1675     1.6KB/s   00:00
id_rsa.pub                                                       100%  400     0.4KB/s   00:00
$ scp -r /tmp/rnmcluster02-ssh-keys/ root@rnmcluster02-node02:~/.ssh
Warning: Permanently added the RSA host key for IP address '172.28.128.7' to the list of known hosts.
root@rnmcluster02-node02's password:
authorized_keys                                                  100%  781     0.8KB/s   00:00
id_rsa                                                           100% 1675     1.6KB/s   00:00
id_rsa.pub                                                       100%  400     0.4KB/s   00:00
$ scp -r /tmp/rnmcluster02-ssh-keys/ root@rnmcluster02-node03:~/.ssh
root@rnmcluster02-node03's password:
authorized_keys                                                  100%  781     0.8KB/s   00:00
id_rsa                                                           100% 1675     1.6KB/s   00:00
id_rsa.pub                                                       100%  400     0.4KB/s   00:00
$ scp -r /tmp/rnmcluster02-ssh-keys/ root@rnmcluster02-node04:~/.ssh
root@rnmcluster02-node04's password:
authorized_keys                                                  100%  781     0.8KB/s   00:00
id_rsa                                                           100% 1675     1.6KB/s   00:00
id_rsa.pub                                                       100%  400     0.4KB/s   00:00

Testing login authenticated through SSH keys

The moment of truth. From your client machine, try to ssh to each of the cluster nodes. If you are prompted for a password, then something is not right – see the troubleshooting section below.

If you put your own public key in authorized_keys when you created it then you don’t need to specify which key to use when connecting because it’ll use your own private key by default:

robin@RNMMBP ~ $ ssh root@rnmcluster02-node01
Last login: Fri Nov 28 17:13:23 2014 from 172.28.128.1



[root@localhost ~]#

There we go – logged in automagically with no password prompt. If we’re using the cluster’s private key (rather than our own) you need to specify it with -i when you connect.

robin@RNMMBP ~ $ ssh -i /tmp/rnmcluster02-ssh-keys/id_rsa root@rnmcluster02-node01
Last login: Fri Nov 28 17:13:23 2014 from 172.28.128.1



[root@localhost ~]#

Troubleshooting SSH key connections

SSH keys are one of the best things in a sysadmin’s toolkit, but when they don’t work can be a bit tricky to sort out. The first thing to check is that on the target machine the authorized_keys file that does all the magic (by listing the ssh keys that are permitted to connect inbound on a host to the given user) is in place:

[root@localhost .ssh]# ls -l ~/.ssh/authorized_keys
-rw-r--r-- 1 root root 775 Nov 30 18:55 /root/.ssh/authorized_keys

If you get this:

[root@localhost .ssh]# ls -l ~/.ssh/authorized_keys
ls: cannot access /root/.ssh/authorized_keys: No such file or directory

then you have a problem.

One possible issue in this specific instance could be that the above pre-canned scp assumes that the user’s .ssh folder doesn’t already exist (since it doesn’t, on brand new servers) and so specifies it as the target name for the whole rnmcluster02-ssh-keys folder. However if it does already exist then it ends up copying the rnmcluster02-ssh-keys folder into the .ssh folder:

[root@localhost .ssh]# ls -lR
.:
total 12
-rw------- 1 root root 1675 Nov 22  2013 id_rsa
-rw-r--r-- 1 root root  394 Nov 22  2013 id_rsa.pub
drwxr-xr-x 2 root root 4096 Nov 30 18:49 rnmcluster02-ssh-keys

./rnmcluster02-ssh-keys:
total 12
-rw-r--r-- 1 root root  775 Nov 30 18:49 authorized_keys
-rw------- 1 root root 1675 Nov 30 18:49 id_rsa
-rw-r--r-- 1 root root  394 Nov 30 18:49 id_rsa.pub
[root@localhost .ssh]#

To fix this simply move the authorized_keys from rnmcluster02-ssh-keys back into .ssh:

[root@localhost .ssh]# mv ~/.ssh/rnmcluster02-ssh-keys/authorized_keys ~/.ssh/

Other frequent causes of problems are file/folder permissions that are too lax on the target user’s .ssh folder (which can be fixed with chmod -R 700 ~/.ssh) or the connecting user’s ssh private key (fix: chmod 600 id_rsa). The latter will show on connection attempts very clearly:

robin@RNMMBP ~ $ ssh -i /tmp/rnmcluster02-ssh-keys/id_rsa root@rnmcluster02-node01
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
@         WARNING: UNPROTECTED PRIVATE KEY FILE!          @
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
Permissions 0777 for '/tmp/rnmcluster02-ssh-keys/id_rsa' are too open.
It is required that your private key files are NOT accessible by others.
This private key will be ignored.
bad permissions: ignore key: /tmp/rnmcluster02-ssh-keys/id_rsa

Another one that has bitten me twice over time – and that eludes the troubleshooting I’ll demonstrate in a moment – is that SELinux gets stroppy about root access using ssh keys. I always just take this as a handy reminder to disable selinux (in /etc/selinux/config, set SELINUX=disabled), having never had cause to leave it enabled. But, if you do need it enabled you’ll need to hit the interwebs to check the exact cause/solution for this problem.

So to troubleshoot ssh key problems in general do two things. Firstly from the client side, specify verbosity (-v for a bit of verbosity, -vvv for most)

ssh -v -i /tmp/rnmcluster02-ssh-keys/id_rsa root@rnmcluster02-node01

You should observe ssh trying to use the private key, and if the server rejects it it’ll fall back to any other ssh private keys it can find, and then password authentication:

[...]
debug1: Offering RSA public key: /tmp/rnmcluster02-ssh-keys/id_rsa
debug1: Authentications that can continue: publickey,gssapi-keyex,gssapi-with-mic,password
debug1: Next authentication method: password

Quite often the problem will be on the server side, so assuming that you can still connect to the server (eg through the physical console, or using password authentication) then go and check /var/log/secure where you’ll see all logs relating to attempted connections. Here’s the log file corresponding to the above client log, where ssh key authentication is attempted but fails, and then password authentication is used to successfully connect:

Nov 30 18:15:05 localhost sshd[13156]: Authentication refused: bad ownership or modes for file /root/.ssh/authorized_keys
Nov 30 18:15:15 localhost sshd[13156]: Accepted password for root from 172.28.128.1 port 59305 ssh2
Nov 30 18:15:15 localhost sshd[13156]: pam_unix(sshd:session): session opened for user root by (uid=0)

Now we can see clearly what the problem is – “bad ownership or modes for file /root/.ssh/authorized_keys”.

The last roll of the troubleshooting dice is to get sshd (the ssh daemon that runs on the host we’re trying to connect to) to issue more verbose logs. You can either set LogLevel DEBUG1 (or DEBUG2, or DEBUG3) in /etc/ssh/sshd_config and restart the ssh daemon (service sshd restart), or you can actually run a (second) ssh daemon from the host with specific logging. This would be appropriate on a multi-user server where you can’t just go changing sshd configuration. To run a second instance of sshd you’d use:

/usr/sbin/sshd -D -d -p 2222

You have to run sshd from an absolute path (you’ll get told this if you try not to). The -D flag stops it running as a daemon and instead runs interactively, so we can see easily all the output from it. -d specifies the debug logging (-dd or -ddd for greater levels of verbosity), and -p 2222 tells sshd to listen on port 2222. Since we’re doing this on top of the existing sshd, we obviously can’t use the default ssh port (22) so pick another port that is available (and not blocked by a firewall).

Now on the client retry the connection, but pointing to the port of the interactive sshd instance:

ssh -v -p 2222 -i /tmp/rnmcluster02-ssh-keys/id_rsa root@rnmcluster02-node01

When you run the command on the client you should get both the client and host machine debug output go crackers for a second, giving you plenty of diagnostics to pore through and analyse the ssh handshake etc to get to the root of the issue.

Hopefully you’ve now sorted your SSH keys, because in the next article we’re going to see how we can use them to run commands against multiple servers at once using pdsh.

Summary

When working with multiple Linux machines I would first and foremost make sure SSH keys are set up in order to ease management through password-less logins.

We’ll see in the next couple of articles some other tools that are useful when working on a cluster:

  • pdsh
  • colmux

I’m interested in what you think – what particular tools or tips do you have for working with a cluster of Linux machines? Leave your answers in the comments below, or tweet them to me at @rmoff.

Categories: BI & Warehousing

Going Beyond MapReduce for Hadoop ETL Pt.3 : Introducing Apache Spark

Rittman Mead Consulting - Tue, 2014-12-09 02:00

In the first two posts in this three part series on going beyond MapReduce for Hadoop ETL, we looked at why MapReduce and Hadoop 1.0 was only really suitable for batch processing, and how the new Apache Tez framework enabled by Apache YARN on the Hadoop 2.0 platform can be swapped-in for MapReduce to improve the performance of existing Pig and Hive scripts. Today though in the final post I want to take a look at Apache Spark, the next-gen compute framework that Cloudera are backing as the long-term successor to MapReduce.

Like Tez, Apache Spark supports DAGs that describe the entire dataflow process, not just individual map and reduce jobs. Like Pig, it has a concept of datasets (Pig’s aliases and relations), but crucially these datasets (RDDs, or “resilient distributed datasets”) can be cached in-memory, fail-back gracefully to disk and can be rebuilt using a graph that says how to reconstruct. With Tez, individual jobs in the DAG can now hand-off their output to the next job in-memory rather than having to stage in HDFS, but Spark uses memory for the actual datasets and is a much better choice for the types of iterative, machine-learning tasks that you tend to do on Hadoop systems. Moreover, Spark has arguably a richer API and when used with Scala, a functional programming-orientated language that uses Java libraries and whose collections framework maps well on to the types of operations you’d want to make use of with dataflow-type applications on a cluster.

Spark can run standalone, on YARN or on other cluster management platforms, and comes with a handy command-line interpreter that you can use to interactively load, filter, analyse and work with RDDs. Cloudera CDH5.2 comes with Spark 1.0.1 and can either be configured standalone or to run on YARN, with Spark as a service added to nodes in the cluster using Cloudera Manager. 

NewImage

So looking back at the Pig example, we create the dataflow using a number of aliases in that case, that we progressively filter, transform, join together and then aggregate to get to the final top ten set of pages from the website logs. Translating that dataflow to Spark we end up with a similar set of RDDs that take our initial set of logs, apply transformations and join the datasets to store the final aggregated output back on HDFS.

NewImage

Spark supports in-memory sharing of data within a single DAG (i.e. RDD to RDD), but also between DAGs running in the same Spark instance. As such, Spark becomes a great framework for doing iterative and cyclic data analysis, and can make much better use of the memory on cluster servers whilst still using disk for overflow data and persistence.

Moreover, Spark powers a number of higher-level tools build on the core Spark engine to provide features like real-time loading and analysis (Spark Streaming), SQL access and integration with Hive (Spark SQL), machine learning (MLib) and so forth. In fact, as well as Hive and Pig being reworked to run on Tez there’s also projects underway to port them both to Spark, though to be honest they’re both at early stages compared to Tez integration and most probably you’ll be using Scala, Java or Python to work with Spark now.

NewImage

So taking the Pig script we had earlier and translating that to the same dataflow in Spark and Scala, we end up with something like this:

package com.cloudera.analyzeblog
import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.SparkContext._
import org.apache.spark.sql.SQLContext
case class accessLogRow(host: String, identity: String, user: String, time: String, request: String, status: String, size: String, referer: String, agent: String)
case class pageRow(host: String, request_page: String, status: String, agent: String)
case class postRow(post_id: String, title: String, post_date: String, post_type: String, author: String, url: String, generated_url: String)
object analyzeBlog {
        def getRequestUrl(s: String): String = {
        try {
                s.split(' ')(1)
        } catch {
                case e: ArrayIndexOutOfBoundsException => { "N/A" }
        }
}
        def main(args: Array[String]) {
val sc = new SparkContext(new SparkConf().setAppName("analyzeBlog"))
val sqlContext = new SQLContext(sc)
import sqlContext._
val raw_logs = "/user/mrittman/rm_logs"
//val rowRegex = """^([0-9.]+)\s([\w.-]+)\s([\w.-]+)\s(\[[^\[\]]+\])\s"((?:[^"]|\")+)"\s(\d{3})\s(\d+|-)\s"((?:[^"]|\")+)"\s"((?:[^"]|\")+)"$""".r
val rowRegex = """^([\d.]+) (\S+) (\S+) \[([\w\d:/]+\s[+\-]\d{4})\] "(.+?)" (\d{3}) ([\d\-]+) "([^"]+)" "([^"]+)".*""".r

val logs_base = sc.textFile(raw_logs) flatMap {
                        case rowRegex(host, identity, user, time, request, status, size, referer, agent) =>
                                Seq(accessLogRow(host, identity, user, time, request, status, size, referer, agent))
                        case _ => Nil
                                }
val logs_base_nobots = logs_base.filter( r => ! r.request.matches(".*(spider|robot|bot|slurp|bot|monitis|Baiduspider|AhrefsBot|EasouSpider|HTTrack|Uptime|FeedFetcher|dummy).*"))

val logs_base_page = logs_base_nobots.map { r =>
  val request = getRequestUrl(r.request)
  val request_formatted = if (request.charAt(request.length-1).toString == "/") request else request.concat("/")
  (r.host, request_formatted, r.status, r.agent)
}

val logs_base_page_schemaRDD = logs_base_page.map(p => pageRow(p._1, p._2, p._3, p._4))

logs_base_page_schemaRDD.registerAsTable("logs_base_page")

val page_count = sql("SELECT request_page, count(*) as hits FROM logs_base_page GROUP BY request_page").registerAsTable("page_count")

val postsLocation = "/user/mrittman/posts.psv"

val posts = sc.textFile(postsLocation).map{ line =>
        val cols=line.split('|')

        postRow(cols(0),cols(1),cols(2),cols(3),cols(4),cols(5),cols(6).concat("/"))
}

posts.registerAsTable("posts")

val pages_and_posts_details = sql("SELECT p.request_page, p.hits, ps.title, ps.author FROM page_count p JOIN posts ps ON p.request_page = ps.generated_url ORDER BY hits DESC LIMIT 10")

pages_and_posts_details.saveAsTextFile("/user/mrittman/top_10_pages_and_author4")

        }
}

I’ll do a code-walkthrough for this Spark application in a future post, but for now note the map and flatMap Scala collection functions used to transform RDDs, and the sql(“…”) function that allows us to register RDDs as tables and then manipulate the contents using SQL, including joining to other RDDs registered as tables. For now though, let’s run the application on the CDH5.2 using YARN and see how long it takes to process the same set of log files (remember, the Pig script on this CDH5.2 cluster took around 5 minutes to run, and the Pig on Tez version on the Hortonworks cluster was around 2.5 minutes:

[mrittman@bdanode1 analyzeBlog]$ spark-submit --class com.cloudera.analyzeblog.analyzeBlog --master yarn target/analyzeblog-0.0.1-SNAPSHOT.jar 
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/opt/cloudera/parcels/CDH-5.2.0-1.cdh5.2.0.p0.36/jars/spark-assembly-1.1.0-cdh5.2.0-hadoop2.5.0-cdh5.2.0.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/opt/cloudera/parcels/CDH-5.2.0-1.cdh5.2.0.p0.36/jars/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
14/12/06 19:18:25 INFO SecurityManager: Changing view acls to: mrittman
14/12/06 19:18:25 INFO SecurityManager: Changing modify acls to: mrittman
...
14/12/06 19:19:41 INFO DAGScheduler: Stage 0 (takeOrdered at basicOperators.scala:171) finished in 3.585 s
14/12/06 19:19:41 INFO SparkContext: Job finished: takeOrdered at basicOperators.scala:171, took 53.591560036 s
14/12/06 19:19:41 INFO SparkContext: Starting job: saveAsTextFile at analyzeBlog.scala:56
14/12/06 19:19:41 INFO DAGScheduler: Got job 1 (saveAsTextFile at analyzeBlog.scala:56) with 1 output partitions (allowLocal=false)
14/12/06 19:19:41 INFO DAGScheduler: Final stage: Stage 3(saveAsTextFile at analyzeBlog.scala:56)
14/12/06 19:19:41 INFO DAGScheduler: Parents of final stage: List()
14/12/06 19:19:41 INFO DAGScheduler: Missing parents: List()
14/12/06 19:19:41 INFO DAGScheduler: Submitting Stage 3 (MappedRDD[15] at saveAsTextFile at analyzeBlog.scala:56), which has no missing parents
14/12/06 19:19:42 INFO MemoryStore: ensureFreeSpace(64080) called with curMem=407084, maxMem=278302556
14/12/06 19:19:42 INFO MemoryStore: Block broadcast_5 stored as values in memory (estimated size 62.6 KB, free 265.0 MB)
14/12/06 19:19:42 INFO MemoryStore: ensureFreeSpace(22386) called with curMem=471164, maxMem=278302556
14/12/06 19:19:42 INFO MemoryStore: Block broadcast_5_piece0 stored as bytes in memory (estimated size 21.9 KB, free 264.9 MB)
14/12/06 19:19:42 INFO BlockManagerInfo: Added broadcast_5_piece0 in memory on bdanode1.rittmandev.com:44486 (size: 21.9 KB, free: 265.3 MB)
14/12/06 19:19:42 INFO BlockManagerMaster: Updated info of block broadcast_5_piece0
14/12/06 19:19:42 INFO DAGScheduler: Submitting 1 missing tasks from Stage 3 (MappedRDD[15] at saveAsTextFile at analyzeBlog.scala:56)
14/12/06 19:19:42 INFO YarnClientClusterScheduler: Adding task set 3.0 with 1 tasks
14/12/06 19:19:42 INFO TaskSetManager: Starting task 0.0 in stage 3.0 (TID 215, bdanode5.rittmandev.com, PROCESS_LOCAL, 3331 bytes)
14/12/06 19:19:42 INFO BlockManagerInfo: Added broadcast_5_piece0 in memory on bdanode5.rittmandev.com:13962 (size: 21.9 KB, free: 530.2 MB)
14/12/06 19:19:42 INFO TaskSetManager: Finished task 0.0 in stage 3.0 (TID 215) in 311 ms on bdanode5.rittmandev.com (1/1)
14/12/06 19:19:42 INFO YarnClientClusterScheduler: Removed TaskSet 3.0, whose tasks have all completed, from pool 
14/12/06 19:19:42 INFO DAGScheduler: Stage 3 (saveAsTextFile at analyzeBlog.scala:56) finished in 0.312 s
14/12/06 19:19:42 INFO SparkContext: Job finished: saveAsTextFile at analyzeBlog.scala:56, took 0.373096676 s

It ran in just over a minute in the end, and most of that was around submitting the job to YARN – not bad. We’ll be covering more of Spark on the blog over the next few weeks including streaming and machine learning examples, and connecting it to ODI and OBIEE via Hive on Spark, and Spark SQL’s own Hive-compatible Thrift server. I’ll also be taking a look at Pig on Spark (or “Spork”…) to see how well that works, and most interestingly how Pig and Hive on Spark compares to running them on Tez – watch this space as they say.

Categories: BI & Warehousing

Six Months an ACE…”Boy, that escalated quickly”

Rittman Mead Consulting - Mon, 2014-12-08 12:34

Boy, that escalated quickly. I mean, my path to becoming an Oracle ACE since joining Rittman Mead!

When I began with Rittman Mead back in March 2012, I wasn’t planning on joining the ACE program. In fact, I really wasn’t sure what an Oracle ACE even was. If you’re not familiar, the Oracle ACE program “highlights excellence within the global Oracle community by recognizing individuals who have demonstrated both technical proficiency and strong credentials as community enthusiasts and advocates”. That’s quite a mouthful! Now let’s find out how I got to this point in my career.

I wrote my first blog post ever shortly after joining Rittman Mead, sharing an innovative way to integrate Oracle Data Integrator and GoldenGate for data warehousing. This lead to submitting my first abstracts to several Oracle User Group conferences. My goal was to challenge myself to grow as a public speaker, something I had never been very good at in the past, and to share my data integration know-how, of course. To my amazement, the first abstract submitted, for UKOUG 2012, was accepted. Then, the other conferences continued to accept my abstracts! During the first couple of years, I kept blogging, speaking, wrote an article for RMOUG SQL>UPDATE Magazine, and joined in on a couple of ODTUG ODI expert webcasts and OTN ArchBeat podcasts over the past couple of years. Before I knew it, I had taken advantage of many knowledge sharing opportunities and built a decent list for my Oracle ACE nomination. I’ve found that networking and sharing your experiences are both wonderful ways to get noticed by other Oracle experts.

BI Forum 2014

So now what is an Oracle ACE to me? I think it is someone knowledgeable about an Oracle product (or products) who enjoys sharing their knowledge and loves contributing to the greater Oracle community. It’s that simple. One thing about the Rittman Mead organization, sharing what we know is a part of the ethos of the company. If you’re an avid reader this blog, you’re well aware that Mark Rittman and others love to share their technical knowledge. We also have an internal system that allows us to share our know-how amongst each other. That’s one of the great benefits of working at Rittman Mead. If I don’t know the answer to a question, I can easily reach out to the entire company and get a response from experts like Mark, James Coyle, Andy Rocha or others within minutes. I love that about this organization!

Well, around March or April timeframe, it was mentioned in a company meeting that if anyone has the goal of joining the ACE program, reach out to one of our current ACEs or ACE Directors and work on a plan to get there (again, a testament to how the RM organization works and thinks). I talked with Stewart Bryson, Oracle ACE Director. After reviewing my contribution to the Oracle community he agreed to submit the nomination for me. Again, to my amazement, and very much my delight, my nomination was accepted as an Oracle ACE! What began as several small goals of improving my writing and speaking skills, sharing technical content, and having an article published, led to the overall end goal (though not known to me at the time) of an Oracle ACE program invitation. I worked hard to earn the nomination, but I also know I was very fortunate to have been surrounded by, and mentored by, some great individuals (and I still am!).

oow14-badge-small

Since joining the ACE program, I’ve gained some additional Twitter followers, a great new ribbon on my conference badges, and some sweet ACE gear – but really it feels like not too much else has changed. Well, I do have an additional group of experts available to help me when I’m in a bind, because that’s what ACEs tend to do. I can now send a question via a tweet or email to someone in the ACE program and I’ll typically get an expert answer. It’s a community of like-minded individuals that love their specific Oracle technology – and love to share knowledge and help others. If this sounds like you, it’s time to work towards that ACE nomination!

Here at Rittman Mead we have several ACEs: myself, Venkat JanakiramanEdelweiss Kammermann, and our newest ACE, Robin Moffatt, as well as an ACE Director, Mark Rittman. And our ACEs don’t just sit back and blog or speak at conferences (not that blogging and speaking isn’t hard work – it is!). We also provide all of the Rittman Mead services offered as consultants: training, consulting services, managed services support – all of it! Which means that when you hire Rittman Mead, the 5 folks in the ACE program are included, along with the other 100+ BI experts throughout the world. If you need some help on your Business Intelligence, Data Integration, or Advanced Analytics project, drop us a line at info@rittmanmead.com and we’ll be happy to have a chat.

Now that I’ve met the goal of Oracle ACE, what’s ahead for me? Well, I’m constantly working to keep up on the latest and greatest from the Oracle Data Integration team, while also consulting full-time. I plan to continue speaking and writing blog posts, but also work towards some additional published content in newsletters and magazines for the Oracle community. And I’m going to keep learning, because education is a never-ending journey. “They’ve done studies, you know. 60 percent of the time, it works every time.” — Brian Fantana 

By the way, if you think Rittman Mead is the type of company you’d like to work for, give us a shout at careers@rittmanmead.com.

Categories: BI & Warehousing

Going Beyond MapReduce for Hadoop ETL Pt.2 : Introducing Apache YARN and Apache Tez

Rittman Mead Consulting - Mon, 2014-12-08 02:00

In the first post in this three part series on going beyond MapReduce for Hadoop ETL, I looked at how a typical Apache Pig script gets compiled into a series of MapReduce jobs, and those MapReduce jobs pass data between themselves by writing intermediate resultsets to disk (HDFS, the Hadoop cluster file system). As a reminder, here’s the Pig script we’re working with:

register /opt/cloudera/parcels/CDH/lib/pig/piggybank.jar
raw_logs = LOAD '/user/mrittman/rm_logs' USING TextLoader AS (line:chararray);
logs_base = FOREACH raw_logs
GENERATE FLATTEN
  (REGEX_EXTRACT_ALL(line,'^(\\S+) (\\S+) (\\S+) \\[([\\w:/]+\\s[+\\-]\\d{4})\\] "(.+?)" (\\S+) (\\S+) "([^"]*)" "([^"]*)"')
)AS
  (remoteAddr: chararray, remoteLogname: chararray, user: chararray,time: chararray, request: chararray, status: chararray, bytes_string: chararray,referrer:chararray,browser: chararray);
logs_base_nobots = FILTER logs_base BY NOT (browser matches '.*(spider|robot|bot|slurp|bot|monitis|Baiduspider|AhrefsBot|EasouSpider|HTTrack|Uptime|FeedFetcher|dummy).*');
logs_base_page = FOREACH logs_base_nobots GENERATE SUBSTRING(time,0,2) as day, SUBSTRING(time,3,6) as month, SUBSTRING(time,7,11) as year, FLATTEN(STRSPLIT(request,' ',5)) AS (method:chararray, request_page:chararray, protocol:chararray), remoteAddr, status;
logs_base_page_cleaned = FILTER logs_base_page BY NOT (SUBSTRING(request_page,0,3) == '/wp' or request_page == '/' or SUBSTRING(request_page,0,7) == '/files/' or SUBSTRING(request_page,0,12) == '/favicon.ico');
logs_base_page_cleaned_by_page = GROUP logs_base_page_cleaned BY request_page;
page_count = FOREACH logs_base_page_cleaned_by_page GENERATE FLATTEN(group) as request_page, COUNT(logs_base_page_cleaned) as hits;
page_count_sorted = ORDER page_count BY hits DESC;
page_count_top_10 = LIMIT page_count_sorted 10;
posts = LOAD '/user/mrittman/posts.csv' USING org.apache.pig.piggybank.storage.CSVExcelStorage() as (post_id:int,title:chararray,post_date:chararray,post_type:chararray,author:chararray,url:chararray,generated_url:chararray);
posts_cleaned = FOREACH posts GENERATE CONCAT(generated_url,'/') as page_url,author as author, title as title;
pages_and_post_details = JOIN page_count by request_page, posts_cleaned by page_url;
pages_and_posts_trim = FOREACH pages_and_post_details GENERATE page_count::request_page as request_page, posts_cleaned::author as author, posts_cleaned::title as title, page_count::hits as hits;
pages_and_posts_sorted = ORDER pages_and_posts_trim BY hits DESC;
pages_and_post_top_10 = LIMIT pages_and_posts_sorted 10;
store pages_and_post_top_10 into 'top_10s/pages';

When you submit the script for execution using the Pig client, it then parses the Pig Latin script, logically optimizes it and then compiles it into MapReduce programs; these programs are then sent in-turn to the Hadoop 1.0 Job Manager which then sends them for execution on the Hadoop cluster – in the case of our Pig script, there’s five MapReduce programs generated in-total.

NewImage

Each one of these MapReduce programs are what’s called “directed acyclic graphs”, or DAGs. A directed acyclic graph is a programming style within distributed computing where processing is broken down into functions that can be run independently of one another, as long as one is not an ancestor of another. In MapReduce terms, this means that all mappers can run independently of each other (and therefore on different nodes in a cluster) and it’s only the reducers that have to wait for their ancestors to finish before they can also start their work independently of the other reducers. It’s a great programming model for processing large amounts of data with fault tolerance across a cluster and it was the key insight that made MapReduce possible and the “big data” systems that we work with today.

NewImage

A Pig script that generates five MapReduce jobs then effectively creates five DAGs, with each one using files (HDFS) to persist and hand-off data between them and JVMs being spun-up for the individual map and reduce jobs. As such, there’s no way for this version of MapReduce and the Hadoop framework it runs on to consider the overall dataflow, and each DAG has to run in isolation.

NewImage

To address this issue, version 2.0 of Hadoop introduced a new feature called Apache YARN, or “yet another resource negotiator”. YARN took on the resource management and job scheduling/monitoring parts of Hadoop and made it so that YARN then effectively became an “operating system” for Hadoop that then allowed frameworks to run on it; initially MapReduce2 (reworked to run on YARN), but since then other ones like Apache Tez, and Apache Spark.

NewImage

YARN also crucially supported frameworks that used DAGs that describe the entire dataflow, not just individual MapReduce jobs, and a new framework that came out of this that used that new capability was Apache Tez. Tez is a generalisation of the MapReduce distributed compute framework that supports these dataflow-style DAGs and runs either MapReduce code unaltered, or has its own API for describing DAGs using vertexes  (logic and resources) or edges (connections). Both Hive and Pig have been ported to run on Tez, and in Pig’s case this means another type of compiler is added to the existing MapReduce one, and Pig scripts executed on Tez can now submit a DAG that encompasses all stages in the dataflow and the reducers and be linked-together via in-memory passing of intermediate steps, rather than having to write those intermediate steps to disk.

NewImage

In practice, what this means is that if your version of Pig or Hive has been updated to also run on Tez, you can run your code unaltered on Tez and typically see a 2-3x performance improvement without any changes to your code or application logic. Hortonworks have been the main Hadoop vendor backing Tez and their new Hortonworks Data Platform 2.2 comes with Tez support, so I took my Pig script and ran it on a five-node cluster to get an initial timing and it took 4 min 17 secs to run, again generating the same five MapReduce jobs that I saw on the Cloudera CDH5 cluster. Then, running it using Tez as the execution engine (pig -x tez analyzeblog.pig) the same script took 2mins 25secs to run, around twice as fast as when we ran it using regular MapReduce.

NewImage

 

And Tez is great for getting existing Pig and Hive scripts to run faster – if your platform supports it, you should use it instead of MapReduce as your execution engine; MapReduce code submitted “as is” will benefit from better YARN container re-use, and Hive and Pig scripts that run on the Tez execution engine can run as a single DAG and use memory, rather than disk, to pass data between jobs in the DAG. For ETL and analytic jobs that you’re creating from new though, Apache Spark is arguably the framework you should look to use instead of Tez, and tomorrow we’ll find out why.

Categories: BI & Warehousing

Going Beyond MapReduce for Hadoop ETL Pt.1 : Why MapReduce Is Only for Batch Processing

Rittman Mead Consulting - Sun, 2014-12-07 08:28

Over the previous few months I’ve been looking at the various ways you can load data into Hadoop, process it and then report on it using Oracle tools. We’ve looked at Apache Hive and how it provides a SQL layer over Hadoop, making it possible for tools like ODI and OBIEE to use their usual SQL set-based process approach to access Hadoop data; later on, we looked at another Hadoop tool, Apache Pig, which provides a more dataflow-type language over Hadoop for when you want to create step-by-step data pipelines for processing data. Under the covers, both Hive and Pig generate Java MapReduce code to actually move data around, with MapReduce then working hand-in-hand with the Hadoop framework to run your jobs in parallel across the cluster.

But MapReduce can be slow; it’s designed for very large datasets and batch processing, with overall analysis tasks broken-down into individual map and reduce tasks that start by reading data off disk, do their thing and then write the intermediate results back to disk again.

NewImage

Whilst this approach means the system is extremely fault-tolerant and effectively infinitely-scalable, this writing to disk of each step in the process means that MapReduce jobs typically take a long time to run and don’t really take advantage of the RAM that’s available in today’s commodity servers. Whilst this is a limitation most early adopters of Hadoop were happy to live with (in exchange for being able to analyse cheaply data on a scale previously unheard of), over the past few years as Hadoop adoption has broadened there’s been a number of initiatives to move Hadoop past it’s batch processing roots and into something more real-time that does more of its processing in-memory. Whilst there are whole bunch of projects and products out there that claim to improve the speed of Hadoop processing and bring in-memory capabilities – Apache Drill, Cloudera Impala, Oracle’s Big Data SQL are just some examples – the two that are probably of most interest to Hadoop customers working in an Oracle environment are called Apache Spark, and Apache Tez. But before we get into the details of Spark, Tez and how they improve over MapReduce, let’s take a look at why MapReduce can be slow.

MapReduce and Hadoop 1.0 – Scalable, Fault-Tolerant, but Aimed at Batch Processing

Going back to MapReduce and what’s now termed “Hadoop 1.0”, MapReduce works on the principle of breaking larger jobs down into lots of smaller ones, with each one running independently and persisting its results back to disk at the end to ensure data doesn’t get lost if a server node breaks down. To take an example, the Apache Pig script below reads in some webserver log files, parses and filters them, aggregates the data and then joins it to another Hadoop dataset before outputting the results to a directory in the HDFS storage layer:

register /opt/cloudera/parcels/CDH/lib/pig/piggybank.jar
raw_logs = LOAD '/user/mrittman/rm_logs' USING TextLoader AS (line:chararray);
logs_base = FOREACH raw_logs
GENERATE FLATTEN
  (REGEX_EXTRACT_ALL(line,'^(\\S+) (\\S+) (\\S+) \\[([\\w:/]+\\s[+\\-]\\d{4})\\] "(.+?)" (\\S+) (\\S+) "([^"]*)" "([^"]*)"')
)AS
  (remoteAddr: chararray, remoteLogname: chararray, user: chararray,time: chararray, request: chararray, status: chararray, bytes_string: chararray,referrer:chararray,browser: chararray);
logs_base_nobots = FILTER logs_base BY NOT (browser matches '.*(spider|robot|bot|slurp|bot|monitis|Baiduspider|AhrefsBot|EasouSpider|HTTrack|Uptime|FeedFetcher|dummy).*');
logs_base_page = FOREACH logs_base_nobots GENERATE SUBSTRING(time,0,2) as day, SUBSTRING(time,3,6) as month, SUBSTRING(time,7,11) as year, FLATTEN(STRSPLIT(request,' ',5)) AS (method:chararray, request_page:chararray, protocol:chararray), remoteAddr, status;
logs_base_page_cleaned = FILTER logs_base_page BY NOT (SUBSTRING(request_page,0,3) == '/wp' or request_page == '/' or SUBSTRING(request_page,0,7) == '/files/' or SUBSTRING(request_page,0,12) == '/favicon.ico');
logs_base_page_cleaned_by_page = GROUP logs_base_page_cleaned BY request_page;
page_count = FOREACH logs_base_page_cleaned_by_page GENERATE FLATTEN(group) as request_page, COUNT(logs_base_page_cleaned) as hits;
page_count_sorted = ORDER page_count BY hits DESC;
page_count_top_10 = LIMIT page_count_sorted 10;
posts = LOAD '/user/mrittman/posts.csv' USING org.apache.pig.piggybank.storage.CSVExcelStorage() as (post_id:int,title:chararray,post_date:chararray,post_type:chararray,author:chararray,url:chararray,generated_url:chararray);
posts_cleaned = FOREACH posts GENERATE CONCAT(generated_url,'/') as page_url,author as author, title as title;
pages_and_post_details = JOIN page_count by request_page, posts_cleaned by page_url;
pages_and_posts_trim = FOREACH pages_and_post_details GENERATE page_count::request_page as request_page, posts_cleaned::author as author, posts_cleaned::title as title, page_count::hits as hits;
pages_and_posts_sorted = ORDER pages_and_posts_trim BY hits DESC;
pages_and_post_top_10 = LIMIT pages_and_posts_sorted 10;
store pages_and_post_top_10 into 'top_10s/pages';

Pig works by defining what are called “relations” or “aliases”, similar to tables in SQL that contain data or pointers to data. You start by loading data into a relation from a file or other source, and then progressively define further relations take that initial dataset and apply filters, use transformations, re-orientate the data or join it to other relations until you’ve arrived at the final set of data you’re looking for. In this example we start with raw log data, parse it, filter out bit and spider activity, project just the columns we’re interested in and then remove further “noise” from the logs, then join it to reference data and finally return the top ten pages over that period based on total hits.

NewImage

Pig uses something called “lazy evaluation”, where relations you define don’t necessarily get created when they’re defined in the script; instead they’re used as a pointer to data and instructions on how to produce it if needed, with the Pig interpreter only materializing a dataset when it absolutely has to (for example, when you ask it to store a dataset on disk or output to console). Moreover, all the steps leading up to the final dataset you’ve requested are considered as a whole, giving Pig the ability to merge steps, miss out steps completely if they’re not actually needed to produce the final output, and otherwise optimize the flow data through the process.

Running the Pig script and then looking at the console from the script running the Grunt command-line interpreter, you can see that five separate MapReduce jobs were generated to load in the data, filter join and transform it, and then produce the output we requested at the end.

JobId                  Maps Reduces Alias                                                                                                               Feature Outputs
job_1417127396023_0145 12   2       logs_base,logs_base_nobots,logs_base_page,logs_base_page_cleaned,logs_base_page_cleaned_by_page,page_count,raw_logs GROUP_BY,COMBINER 
job_1417127396023_0146 2    1       pages_and_post_details,pages_and_posts_trim,posts,posts_cleaned                                                     HASH_JOIN 
job_1417127396023_0147 1    1       pages_and_posts_sorted                                                                                              SAMPLER 
job_1417127396023_0148 1    1       pages_and_posts_sorted                                                                                              ORDER_BY,COMBINER 
job_1417127396023_0149 1    1       pages_and_posts_sorted                                                                                              hdfs://bdanode1....pages2,

Pig generated five separate MapReduce jobs that loaded, parsed, filtered, aggregated and joined the datasets as part of an overall data “pipeline”, with the intermediate results staged to disk before the next MapReduce job took over. On my six-node CDH5.2 VM cluster it took just over five minutes to load, process and aggregate 5m records from our site’s webserver.

NewImage

Now the advantage of this approach is that its more or less infinitely scalable and certainly resilient, but whilst Pig can look at your overall dataflow “graph” and come up with an optimal efficient way to get to your end result, MapReduce treats every step as atomic and separate and insists on writing every intermediate step to disk before moving on.

What this means in-practice is that ETL routines that use Pig, Hive and MapReduce whilst scaling well, never really get to the point where you can run them as micro-batches or in real-time. For that type of scenario we need to look at moving away from MapReduce and breaking the link between Hadoop (the platform, the cluster management and resource handling part) and the processing that runs on it, so that we can run alternative execution engines on the Hadoop platform such as Apache Tez, which we’ll cover in tomorrow’s post.

Categories: BI & Warehousing

VirtualBox and Windows driver verifier

Amardeep Sidhu - Wed, 2014-12-03 04:34

I was troubleshooting some Windows hangs on my Desktop system running Windows 8 and enabled driver verifier. Today when I tried to start VirtualBox it failed with this error message.

Failed to load VMMR0.r0 (VERR_LDR_MISMATCH_NATIVE)

Most of the online forums were asking to reinstall VirtualBox to fix the issue. But one of the thread mentioned that it was being caused by Windows Driver Verifier. I disabled it, restarted Windows and VirtualBox worked like a charm. Didn’t have time to do more research as i quickly wanted to test something. May be we can skip some particular stuff from Driver Verifier and VirutalBox can then work.

Categories: BI & Warehousing

Data Warehouse for Big Data: Scale-Up vs. Scale-Out

Dylan Wan - Thu, 2014-01-02 15:33

Found a very good paper: http://research.microsoft.com/pubs/204499/a20-appuswamy.pdf


This paper discuss if it is a right approach of using Hadoop as the analytics infrastructure.


It is hard to argue with the industry trend.  However, Hadoop is not
new any more.  It is time for people to calm down and rethink about the
real benefits.

Categories: BI & Warehousing