[OpenAFS] improving cache partition performance

Harald Barth haba@kth.se
Mon, 29 Aug 2011 18:33:44 +0200 (CEST)

> % fs getcacheparms
> AFS using 145647975 of the cache's available 212484585 1K byte blocks.

So why are you "only" at 145 of 212 GB? Either your cache really is
bigger than your working set since boot or you have hit the number of
cache entries before the cache size is exhausted.

> I'm already using RAID 0, so I'm already striping. 

Ah, sorry, when I read RAID, I thought RAID5.

> I'll have to do some benchmarking with ext2.

I will be faster that ext3. The question is how much, which probably
depends on your specific load. I have not made any recent experiments
with other file systems (for example ext4, xfs) as cache yet, I just
know that reiserfs killed my files back in the days... I use xfs as
server (/vicep*/) file system.


PS: Some years ago I did some benchmarking of AFS cache performance. I
was astonished how much slower ext2 on ramdisk was compared to
memcache. So the question is, how fast could a AFS disk cache be if it
did not needed to bother with a file system and could use swap?