[Lustre-discuss] Slow read performance after ~700MB-2GB

[Jason Rappleye]

On Jun 9, 2010, at 1:51 PM, Jason Rappleye wrote:

>
> I haven't tried to reproduce this problem with a 1.8.2 client yet, but
> I should be able to sneak in a test this week and see if it works as
> expected.

Same behavior with what is essentially a 1.8.2.0-24chaos client  
(changes to it are inconsequential) and SLES 10 SP2 kernel  
2.6.16.60-0.42.5 + some local patches. Those patches *should* be  
inconsequential as well - mostly changes to the OOM handler, IIRC  
(which isn't an issue here).

Our vendor already has a case on this - I'll go ahead and have them  
file a BZ.

Thanks,

j

--
Jason Rappleye
System Administrator
NASA Advanced Supercomputing Division
NASA Ames Research Center
Moffett Field, CA 94035