Re: testing memory parameters

From: DA Morgan <damorgan_at_psoug.org>
Date: Thu, 20 Mar 2008 19:54:57 -0700
Message-ID: <1206068094.775903@bubbleator.drizzle.com>


Serge Rielau wrote:

> DA Morgan wrote:
>> Serge Rielau wrote:
>>> DA Morgan wrote:
>>>> hpuxrac wrote:

>>>>> On Mar 18, 4:24 pm, DA Morgan <damor..._at_psoug.org> wrote:
>>>>>
>>>>> snip
>>>>>
>>>>>>>> Because, in DB2, you don't have sufficient instrumentation to look
>>>>>>>> inside and see what is happening.
>>>>>>> Ah, Daniel, you never fail to live up to my expectation of taking a
>>>>>>> cheap shot wherever you can. What would you know about DB2's
>>>>>>> state of
>>>>>>> perf tuning tools...
>>>>>>> But I digress. I got the answers I needed.
>>>>>>> Thanks Ana
>>>>>>> Serge
>>>>>> Perhaps I was wrong ... please provide me the link to the DB2
>>>>>> equivalent
>>>>>> to ASH, AWR, and the wait interface.
>>>>>
>>>>> Why don't you take some time and read the DB2 documentation?
>>>>>
>>>>> Probably you can start with some searches on google ... let's see
>>>>> there is the DB2 Technical Library, IBM Redbooks, stuff like the DB2
>>>>> Tuning Database Performance manual ... etc. Doing a little legwork
>>>>> might be good for you.
>>>>>
>>>>> Of course Oracle offers free access to all the stuff like ASH, AWR,
>>>>> and the wait interface thru OEM or Grid Control ... well at least if
>>>>> you pay extra licensing fee's for the options packs and enterprise
>>>>> edition.
>>>>>
>>>>> Wonder if many oracle customer's skip paying for these options and use
>>>>> 3rd party tools like Quest or BMC etc? Wonder if any other 3rd
>>>>> parties do the same thing for DB2 ... why surprise I guess they do!
>>>>>
>>>>> Looks like you have a bunch of homework to do!
>>>>
>>>> One of the first thing you learn in law school is to not ask a
>>>> question, in court, to which you don't already know the answer.
>>>> Let Serge respond if he wishes.
>>> One of the first things you learn in law-school is that the one 
>>> making the allegation is the one having to prove they are correct.
>>> "Because, in DB2, you don't have sufficient instrumentation to look
>>> inside and see what is happening."
>>> The onus is not on me. Aside this is an Oracle group. I asked a 
>>> question. I got a reasonanble answer from anyone but you.
>>>
>>>> But while we're at it I will clarify the request ... DB2 on LUX not
>>>> DB2 on the mainframe.
>>> What does Luxemburg have to do with this?
>>>
>>> Cheers
>>> Serge
>>
>> Two times you could have provided the name of that equivalent technology
>> built into your product so that people could look it up and see. Twice
>> you have dodged. Here's a third opportunity to set the record straight
>> if you think I am misrepresenting the capabilities of DB2 on Linux UNix.
> If you ask NICELY in c.d.ibm-db2 AND you EXPLAIN these acronyms you're 
> throwing around mean (I'm the SQL guy, remember - that's why I ask the 
> stupid questions about hit ratios) then I'm sure someone will point you 
> to DB2 technologies.
> Oh and it will help if you state whether they are free or add-on's so 
> the net is cast appropriately.
> 
> Cheers
> Serge

My apology Serge. I thought by now you'd have devoured my website and every public doc at tahiti.

I am not familiar with any technology inside of any current Oracle database competitor that has even a fraction of Oracle's built in instrumentation.

These links may help:

http://www.pythian.com/blogs/275/using-the-oracle-wait-interface-to-troubleshoot-io-issues
http://www.orafaq.com/maillist/oracle-l/2002/06/15/1187.htm
http://download.oracle.com/docs/cd/B19306_01/server.102/b14211/autostat.htm#PFGRF02601
http://download.oracle.com/docs/cd/B19306_01/server.102/b14211/autostat.htm#i35568
http://www.freelists.org/archives/oracle-l/05-2004/msg01009.html

Google will bring you a few hundred thousand more. <g>

-- 
Daniel A. Morgan
Oracle Ace Director & Instructor
University of Washington
damorgan_at_x.washington.edu (replace x with u to respond)
Puget Sound Oracle Users Group
www.psoug.org
Received on Thu Mar 20 2008 - 21:54:57 CDT

Original text of this message