On Thu, 8 Mar 2007, Anand Avati wrote:
Brent, glusterfs (client) memory usage increase with metadata operations is expected, it should drop after some idletime (after the kernel wishes to forget the correspending inodes),
I thought that could be, but I'm not sure. I left the filesystem idle all night, and I still have (from "ps auxww"): root 3047 4.7 16.7 674668 673444 ? Ss Mar07 49:26 [glusterfsd] root 3068 4.9 2.5 103072 101584 ? Ss Mar07 51:29 [glusterfs]
though the glusterfsd increase looks suspicious.
Agreed.
By the way, have you been trying io-threads on server or client or both?
Well, I didn't know I could even do a performance translator on the server (glusterfsd)! All my performance translator tests are with the client spec only. My server specs only contain storage/posix and protocol/server. Client and server both are running on the same node(s).
Are performance translators recommended/required for both client and server processes? Can you do the cluster translators on the server-side, too?
Thanks for the notice. (we are looking into read-ahead fixes currently)
Cool beans. Thanks, Brent
