[Gluster-devel] Unfair scheduling in unify/AFR
avati at zresearch.com
Wed Nov 21 18:58:08 UTC 2007
> > You might be
> > interested in the new booster module which is on the way which will
> > short cut the IO path directly to the server from the application's
> > address space (via LD_PRELOAD). You might also want to try the next few
> > TLA commits which are on the way, which will help parallel file reads
> > going via fuse.
> I have tried the booster module, it does the same. Please let me know if
> there are new commits that may help.
There have been new commits which should (theoretically atleast) improve
read performance while read-ahead loaded in the client. Also you should try
loading booster on the server side so that I/O happens from client's address
space directly to the server (bypassing kernel/fuse/glusterfs client). The
configuration is a bit tricky and yet to be documented, but our tests so far
have shown good results so far.
Thanks in advance,
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
It always takes longer than you expect, even when you take into account
-- Hofstadter's Law
More information about the Gluster-devel