[OpenAFS] humungous disk caches

Bryan Bayerdorffer bryan.bayerdorffer@spd.analog.com, bryan.bayerdorffer@spd.analog.com
Wed, 12 Mar 2003 17:47:43 -0600


Derrick J Brashear wrote:
>>How about ext3?  Does the journaling get in the way at all?  Is ext2 the only
>>safe option?
> ext3 is fine.
> 
> reiserfs will blow up now, or later.

Ok, ext3 it is.  So what's the practical upper limit on the size of the disk
cache?  Switching to ext3 didn't affect the apparent 16GB limit, beyond which
the first access to /afs will hang.  Nothing in the kernel log. I haven't yet
straced, but I will if there isn't an easy answer to this.

Our apps are characterized by very large files that are read often and modified
only rarely, so I'd like to see if a large disk cache (128GB) will improve
performance.

-- 
    .. ..-. ..- -.-. .- -. .-. . .- -.. - .... .. ... --. . - .- .-.. .. ..-. .
Bryan Bayerdorffer       bryan@meatspace.net              bryan@spd.analog.com
                      (Wit's End Computation Center)           (Analog Devices)

"Sing, O goddess, the anger of Achilles son of Peleus, that brought countless
ills upon the Achaeans."  -- Homer

"Donuts---is there anything they can't do?"  -- Homer