[OpenAFS] openafs performace problems

Derrick Brashear shadow@gmail.com
Tue, 6 Jan 2009 17:05:53 -0500

>>> Why would attribute caching make the test be an apples - oranges
>>> comparison?
>> Because you have no cache coherancy on NFS to verify that data is propogated
>> out and seen simultaneously on multiple clients
>> (V2 & 3)  whereas with OpenAFS that cache coherancy is there.
> If your existing application is working fine on NFS with attribute
> caching you should _not_ turn it off when you are comparing to AFS.
> The application can clearly work with the attribute cache. (The
> attribute cache is normally not a problem)
> In fact the test would be misleading if you did turn the cache off.

Everything's relative. I won't argue you should turn it off, just that
"it's working fine" is relative. Everything works fine right up until
it doesn't when you haven't changed anything, and at that point, maybe
what was fine before now isn't.

It's all subjective. You have to decide what's right for you, and then
benchmark the conditions you believe work for you. Maybe those change.
If so, hopefully you aren't locked in if your previous choice is now
no longer best.