Re: Also seeing 2 x quad-core system slower that 2 x dual core

Just as a followup to this - I soent some time going through all
the suggestions and advice that people gave me regarding this problem.
It turns out that the newer servers shipped by HP have different cache
settings to the older ones on their RAID controllers, plus I get very
different results from my benchmarks depending on how long the machines
have been booted for and what activity has occurred on them (probably due to
things ending up in cache).

What settings are there on the cache? I have a DL 380 G5 with 2 x
dual-core woodcrest and 512 MB BB cache running as a db-server
(postgresql) and two DL 360 G5 with with 2 x 4-core and 256 MB BB
cache running as web-servers.

Upshot - if the machines are configured identically, and an identical install
is made and an identical test doen then we get identical performance as
expected.

Part of the reason for posting this though is that a lot of people have bbeen
worrying about 8x CPU performance, and this thread won't have helped. So
I wanted to say that now I am convinced that (for my workload) these machines
are fine. To the point where I have installed 7.0-BETA4 on the ten new
8 core servers for a very large load on th webfarm this morning. I'm pleased
toio say that it went off perfectly, the servers took the load and we had
no problems at all. We are running CGI scripts against mySQL under apache22
basically - which is a pretty common thing to do. Ia m using ULE and tthe
amd64 version of the OS.

I will second this even though I do not have hard facts to base this
assumption on. I have - as mentioned above - two web-servers running
7.0 beta 2 on a 8-core and is satisfied with the performance. But my
only comparison are two 2 x dual-core opterons (4-way) where the
8-core handles approx. twice as much load. But maby I should expect
2.5 as much performance since the quad-core is a newer cpu and the
opterons are getting two years old.

I also upgraded our db-server from 6.2 RC1 to 7.0 beta 3 on our DL 380
G5 (4-way) because the server was getting constrained on ressources. I
could either replace my 10K rpm drives (in raid 1+0) with 15K ditto
which would require a downtime which we could not afford at this time
or I could upgrade from 6.2 to 7.0. The upgrade took 10 min. Kernel
and userland had all ready been compiled on the web-server. The
upgrade was worthwhile and the db-server is performing better. On
Sunday I will have some load-numbers and post them on Monday.

--
regards
Claus

When lenity and cruelty play for a kingdom,
the gentlest gamester is the soonest winner.

Shakespeare
_______________________________________________
freebsd-stable@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • DNS poisoning of naive caches, bigred.com search engine
    ... Mydomain.com's servers give different addresses for NS.REFRACT.COM, ... cache the glue from the parents, in this case, the gTLD servers. ... A DNS cache, following delegations from the roots, and the gTLD servers, for ... _iterative_ queries made to one of these poison servers. ...
    (Incidents)
  • Sudden and massive page cache eviction
    ... page cache on some of our servers for reasons that we cannot explain. ... buffer cache to deal with incoming request load. ... We have correlated the incidence of these page eviction with higher ...
    (Linux-Kernel)
  • Re: Sudden and massive page cache eviction
    ... page cache on some of our servers for reasons that we cannot explain. ... buffer cache to deal with incoming request load. ... We have correlated the incidence of these page eviction with higher ...
    (Linux-Kernel)
  • Re: AS2005 ... what is wrong with it?
    ... the question is not cold cache and warm cache during query time. ... stop both servers 32bits and 64bits ... the OLAP database. ...
    (microsoft.public.sqlserver.olap)
  • Re: building on your own a large data storage ...
    ... Which means we should not consider SATA drives for enterprise/ ... that it could not handle in its own write-back cache it might matter some). ... I think google probably uses caches aggreesively. ... way google sanitized off their servers)), ...
    (comp.os.linux.hardware)