On Wed, 16 Jul 2008, Ben Mok wrote:
Hello ,
I did the test to generate 100,000 files on glusterfs, each file is 4k, it
need to take 150m27s . I did the same test in local disk and samba share,
just 1m14s and 4m35s. Can you tell me how to improve the performance , since
I want to use glusterfs to handle a large amount of files.
Thank you very much !!
Ben
FYI, the 1.4 branch is much more efficient handling metadata operations
(switched from ASCII protocol to binary), so it is significantly faster in
situations involving large numbers of small files. It's not in an
official release yet, unfortunately, but it does work nicely.
Perhaps the GlusterFS developers will consider doing an intermediate
release with this improvement or make it an option in 1.3 (probably too
big a change, though, and would be difficult reworking the code to make it
optional), as the first official 1.4 release is probably still a ways
away...
Thanks,
Brent
PS The developers are also working towards eliminating the namespace
requirement for unify/stripe; this could be a significant improvement, as
well.
