I can certainly attest to some weirdness being present in glibc2.1. I
will be upgrading my RH6.0 MySQL server to dual-CPU tomorrow (Tuesday
Jun.29) so I should be able to find out fairly quickly if I encounter a
bottleneck.
I've only had MySQL running on the box for two days; it was on an
OpenBSD platform last week. I never ran benchmarks there - the box was
too slow to bother :-)
-Adam Thompson
<athompso@stripped>
Tom Cunningham wrote:
> The old machine (the one that performs better under heavy hits) is running
> Linux 2.2.5. The new machine is running Linux 2.2.10ac4. The bug seems to
> be in the glibc version, because we've tried 2.2.5 on the new machine, and
> it didn't help performance - the old machine still outperformed the new
> one even though they were both running the same kernel (2.2.5).
>
> The Usenet post also seemed to point a finger at glibc2.1 :
>
>
> http://x23.deja.com/getdoc.xp?AN=474851996&search=thread&CONTEXT=930262832.17345
> 41401&HIT_CONTEXT=930262265.1662714038&HIT_NUM=1&hitnum=1
> Subject : performance bug in glibc2.1 linux 2.2?
> Date: 1999/05/06
>
> Is anyone else that is using Red Hat 6.0 willing to check this out, or has
> anyone seen this happen? I really don't want to retreat to an earlier
> glibc library if I don't have to.
--
Adam Thompson, MCNE, MCSE, CWT, A+
Vice-President / Chief Technology Officer, Commerce Design Inc.
<athompso@stripped>
tel: (204) 942-1648, fax: (204) 989-8080, cell: (204) 782-6198
Attachment: [text/x-vcard] Card for Adam Thompson athompso.vcf
Attachment: [text/x-vcard] Card for Adam Thompson athompso.vcf
Attachment: [application/x-pkcs7-signature] S/MIME Cryptographic Signature smime.p7s
Attachment: [application/x-pkcs7-signature] S/MIME Cryptographic Signature smime.p7s
