Re: Number of CPUs vs. Speed of CPUs

My view:

if you have a lot of contention: few fast CPUs if you have virtually no contention: few fast CPUs or large number of slower CPUs.

if you have a lot of contention and a large number of slow CPUs, your contention will become worse (basically it will take longer to execute the synchronized code and with more cpu's there is more chance for sessions spinning on a latch ).

Anjo.

K Gopalakrishnan wrote:

> Bruce:
>
> You got the point. We have discussed that some time back (Steve, Jonathan,
> and Adrian Cockcroft) and the result was:
>
> Few Faster CPUs are better for Performance, More slower CPUs are better for
> Manageability and fault tolerance.
> Steve Adams and Adrian are with few faster CPUS and Jonathan supports more
> slower CPUs.
>
> Best Regards,
> K Gopalakrishnan
> Bangalore, INDIA
>
> -----Original Message-----
> Bruce (CALBBAY)
> Sent: Thursday, April 18, 2002 4:33 PM
> To: Multiple recipients of list ORACLE-L
>
> Lisa,
>
> These aren't my points or ideas but below are some emails from Steve Adams
> that I have kept that may be of use.
>
> Also, I don't know for sure, but the 2 books (and or their authors:-) "
> Practical Oracle8i by Jonathan Lewis" and "Scaling Oracle8i by James Morle"
> may well have some very useful information on this topic.
>
> HTH,
> Bruce Reardon
>
> -----Original Message-----
> Sent: Thursday, 1 March 2001 17:46
> To: Ixora Answers
>
> Hi All,
>
> Here is another follow-up to the February "Ixora News" item "Buying Time" at
> http://www.ixora.com.au/newsletter/2001_02.htm#memory_access.
> One of our subscribers asked Adrian Cockcroft, the author of "Sun
> Performance
> and Tuning" at http://www.amazon.com/exec/obidos/ASIN/0130952494/ixora for a
> second opinion.
> Here is Adrian's response ...
>
> Few faster CPUs is a little more efficient and has lower response time.
>
> More slow CPUs gives a flatter more predictable response time curve
> as the load increases so you can run nearer 100% busy.
>
> Read any book on queuing theory, especially Neil Gunther's
> "Practical Performance Analyst" at
> http://www.amazon.com/exec/obidos/ASIN/059512674X/ixora.
>
> See "Performance Dynamics" at http://www.perfdynamics.com/ for more on Neil
> Gunther's work, and "Performance School"
> at http://www.perfskool.com/ for related downloads.
>
> @ Regards,
> @ Steve Adams
>
> -----Original Message-----
> Sent: Wednesday, 28 February 2001 15:23
> To: Ixora Answers
>
> Hi All,
>
> The recent Ixora News item "Buying Time" at
> http://www.ixora.com.au/newsletter/2001_02.htm#memory_access has provoked
> some
> discussion on the question of whether to buy fewer faster CPUs or more
> slower
> ones.
>
> It has been pointed out that there is one case in which it might be better
> to
> buy more slower CPUs. If an application has a number of highly CPU intensive
> processes (or threads) greater than or equal to the number of CPUs
> available,
> then the CPU usage will be unacceptably high. High CPU usage increases
> context
> switching overheads and introduces a risk of latch contention. If for some
> reason the CPU intensive processes cannot be tamed, then buying a greater
> number
> of slower CPUs may give better overall performance, despite the reduced CPU
> capacity and scalability.
>
> @ Regards,
> @ Steve Adams
>
> -----Original Message-----
> Sent: Friday, 19 April 2002 1:58
>
> We are in the process of sizing a new server for multiple Oracle instances.
> What factors are useful as input in determining how many CPUs and the
> relative speed of them? For example, do we want fewer, faster CPUs or do we
> want more, slower CPUs? Are there any good guidelines to determine what the
> number of CPUs should be?
>
> Thanks in advance -
> Lisa
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: Reardon, Bruce (CALBBAY)
> INET: Bruce.Reardon_at_comalco.riotinto.com.au
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).
>
> --
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> --
> Author: K Gopalakrishnan
> INET: kaygopal_at_yahoo.com
>
> Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> San Diego, California -- Public Internet access / Mailing Lists
> --------------------------------------------------------------------
> To REMOVE yourself from this mailing list, send an E-Mail message
> to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> the message BODY, include a line containing: UNSUB ORACLE-L
> (or the name of mailing list you want to be removed from). You may
> also send the HELP command for other information (like subscribing).

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Anjo Kolk
  INET: anjo_at_oraperf.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
the message BODY, include a line containing: UNSUB ORACLE-L
(or the name of mailing list you want to be removed from).  You may
also send the HELP command for other information (like subscribing).