[OpenAFS] cache performance

[Nickolai Zeldovich]

On Thu, 24 Oct 2002, Phil.Moore@morganstanley.com wrote:

> NOATIME!?  Uh, forgive my ignorance, but I take it this means that
> access times are no longer maintained on the V files in the cache?

Well, I thought this was the behavior for a long time now, but since
you mention it, hm...  It looks like in 3.5p2, Transarc added noatime
code to the Solaris port (osi_DisableAtimes() in SOLARIS/osi_file.c),
and to Linux (in osi_UFSOpen() in LINUX/osi_file.c).  But something
seems to be still updating the atimes, as I'm seeing recent atimes
on most V files in my AFS client caches.

So, although there's code that claims to disable atimes, empirical
evidence suggests that it's broken, and atimes are being updated
anyway (not a big surprise for Transarc/AFS code).  Given that you
are relying on atimes, I'm a bit hesitant to "fix" it, esp. lacking
good performance reasons to do so.

> The reason we have this code is that we analyze the contents of ALL of
> our clients caches (yes, I'm not joking, we really do), and procude
> enterprise wide reports on who is accessing what volumes.  This has
> proven very useful.

Would a combination of cmdebug and server-side logging be a reasonable
alternative?

-- kolya