Re: question on SQL queries

From: Thomas Kyte <tkyte_at_us.oracle.com>
Date: 2000/01/11
Message-ID: <ubpm7s400mtok1pa7ec3aqd3s7cc90jl4g_at_4ax.com>#1/1


A copy of this was sent to "Sybrand Bakker" <postmaster_at_sybrandb.demon.nl> (if that email address didn't require changing) On Tue, 11 Jan 2000 17:22:47 +0100, you wrote:

>Within Oracle, this is not possible.

Actually with OPS (parallel server) and PQ (parallel query) this is possible and its done for you.

<quote>
Oracle Parallel Query on Oracle Parallel Server

Oracle Parallel Server provides the framework for the Parallel Query Option to work between nodes. The Oracle Parallel Query behaves the same way in Oracle with or without the Parallel Server Option. The only difference is that Oracle Parallel Server enables Oracle Parallel Query to ship queries between nodes so that multiple nodes can execute on behalf of a single query. Here, the server breaks the query up it into smaller operations that run against a common database which resides on shared disks. Because it is performed by the server, this parallelism can occur at a low level of server operation, rather than at an external SQL level.
</quote>

All the software to coordinate the result sets and such is part of the server -- you would just submit a query to the server and it would split it up amongst the available nodes...

>It is possible though to use Parallel Query. In this situation you have
>multiple CPUs in one system, and you can use them to distribute parts of
>your execution paths over different CPUs.
>
>Hth,
>
>--
>Sybrand Bakker, Oracle DBA
>Jay Jayaprasad <jayj_at_netscape.net> wrote in message
>news:387B4FBC.56F2C386_at_netscape.net...
>> I am writing an application where we need to analyze large amounts of
>> data. We would like to split up the processing into multiple smaller
>> chunks of rows that are executed on different machines. Each machine
>> should be able to hook up to the database independently and pick up a
>> chunk that is different from one that another machine is getting. One
>> approach would be for each machine to independently execute the same
>> select query and then for the first machine to process rows 1 to 1000,
>> the second machine to process rows 1001 to 2000 etc. This approach would
>> work only if it is guaranteed that the same select query that when
>> executed at different times on the same data would always return the
>> results in the same order. Is this a valid assumption for any Oracle
>> server (including parallel server) and any arbitrarily complex select
>> query? Also, are there other ways of solving this?
>>
>> Thanks,
>> Jay Jayaprasad
>> SPSS Inc.
>>

-- 
See http://osi.oracle.com/~tkyte/ for my columns 'Digging-in to Oracle8i'...
Current article is "Part I of V, Autonomous Transactions" updated June 21'st
 
Thomas Kyte                   tkyte_at_us.oracle.com
Oracle Service Industries     Reston, VA   USA

Opinions are mine and do not necessarily reflect those of Oracle Corporation
Received on Tue Jan 11 2000 - 00:00:00 CET

Original text of this message