[OpenAFS] Re: Expected performance

[Andy Cobaugh]

On 2011-05-19 at 14:19, Andrew Deason ( adeason@sinenomine.net ) said:
> On Thu, 19 May 2011 14:57:16 -0400 (EDT)
> Andy Cobaugh <phalenor@gmail.com> wrote:
>
>> You can certainly get close if your disk for the disk cache is fast
>> enough. I've seen close to 80MB/s with 15K SAS under ideal conditions.
>>
>> Re: client and server on the same machine - I've seen that actually
>> result in lower performance. When you take the physical network out of
>> the mix, Rx starts limiting you as a function of CPU usage it seems.
>
> ...or it's that you're writing to the same disk twice as much. If the
> cache and /vicepX are on the same disk, it seems pretty intuitive that
> it's going to be slower.

It was with memcache.

Just ran some quick tests yesterday to confirm what I saw before.

Here I have two different clients, 'mal' and 'badger'. badger has a 
fileserver. There is another fileserver, fs8, which serves the purpose of 
showing maximum client performance (fs8 is our biggest and fastest 
fileserver currently).

Clients on both mal and badger have essentially the same config, using a 
655360 byte memcache.

iozone was used in all tests.

client -> server

mal -> fs8:
http://www.bx.psu.edu/~phalenor/afs_performance_results/mal.bx.psu.edu-201105121302/

mal -> badger:
http://www.bx.psu.edu/~phalenor/afs_performance_results/mal.bx.psu.edu-201105191536/

badger -> fs8:
http://www.bx.psu.edu/~phalenor/afs_performance_results/badger.bx.psu.edu-201105191708/

badger -> badger:
http://www.bx.psu.edu/~phalenor/afs_performance_results/badger.bx.psu.edu-201105191618/

mal and badger are slightly different hardware, but the tests above show 
that we get very similar performance between all server and client 
combinations except the case where client and server are on the same 
machine.

Maybe I'm missing something here?

--andy