[OpenAFS-devel] good and bad performance of memory cache

Lyle Seaman lws@o-o.yi.org
Wed, 05 Feb 2003 19:03:13 -0500


> Lyle will probably chime in here, but my recollection is that memcache
> is using an array-based implementation of the cache instead of an hash;=

> thus searching is O(n) instead of O(log n).  =


I'm not sure which hash you're referring to, but that doesn't sound right=
 to =

me.
There were some heuristics in the cache manager for filesize > cache size=
, =

perhaps they're only present in the disk cache code and not the mem cache=
 code.

I suspect that if you look at network traffic, or server statistics, you'=
ll =

find that the server is being hit harder in the memcache case.  No?

=2E..apropros of an earlier message:

mountain-dew~/capsule> echo 100001/8 | bc
12500