Re: CPU_COUNT
From: Shakespeare <whatsin_at_xs4all.nl>
Date: Mon, 07 Sep 2009 10:20:02 +0200
Message-ID: <4aa4c233$0$83238$e4fe514c_at_news.xs4all.nl>
sybrandb_at_hccnet.nl schreef:
> On Fri, 4 Sep 2009 12:04:51 -0700 (PDT), Sue <sueschoch_at_gmail.com>
> wrote:
>
>
> I don't think so. The math is quite simple.
> You have 4 cpu's with 4 cores, so CPU_COUNT will be 16. (As you see
> above).
> If you set CPU_COUNT to 1, you will use 1 core of 1 cpu.
> Not that it matters as you do not pay for the number of cores that you
> use, you pay for the number of cores in the server.
>
Date: Mon, 07 Sep 2009 10:20:02 +0200
Message-ID: <4aa4c233$0$83238$e4fe514c_at_news.xs4all.nl>
sybrandb_at_hccnet.nl schreef:
> On Fri, 4 Sep 2009 12:04:51 -0700 (PDT), Sue <sueschoch_at_gmail.com>
> wrote:
>
>> SQL> select * from dba_cpu_usage_statistics; >> DBID VERSION TIMESTAMP CPU_COUNT >> CPU_CORE_COUNT CPU_SOCKET_COUNT >> ---------- ----------------- ------------------- ---------- >> -------------- ---------------- >> 1055427977 11.1.0.7.0 08/22/2009 22:09:14 >> 32 16 4 >> 1055427977 11.1.0.7.0 08/31/2009 22:41:41 >> 1 16 4 >> >> This is actually a 4 CPU machine with 4 cores per CPU. Thanks for >> any help. I've been searching metalink but not finding what I'm >> looking for.
>
> I don't think so. The math is quite simple.
> You have 4 cpu's with 4 cores, so CPU_COUNT will be 16. (As you see
> above).
> If you set CPU_COUNT to 1, you will use 1 core of 1 cpu.
> Not that it matters as you do not pay for the number of cores that you
> use, you pay for the number of cores in the server.
>
Not with hard-partitioning.
Shakespeare Received on Mon Sep 07 2009 - 03:20:02 CDT