[Gluster-devel] Performance issue

Anand Avati avati at zresearch.com
Thu Jul 17 13:58:13 UTC 2008


Ben,
 the qa are internal release to our QA. Please use the tla checkout. the
hash translator is not yet in the mainline. migrating from 1.3 to 1.4
current is straight forward, no changes. We plan to bring in option name
cleanup before the first public release of 1.4 though.

avati

2008/7/17 Ben Mok <benmok at powerallnetworks.com>:

>  Hi Avati and Brent,
>
>
>
> Thank you for your information! I have several questions about 1.4
>
> 1.       the last version is 1.4.0qa29, do new translator hash is
> available in this version?
>
> 2.       I just installed 1.4,  can I use old conf files which are used in
> 1.3 9 ? when I use old conf on client side, it can't start, log showed:
>
>
>
> 2008-07-17 15:33:38 D [unify.c:4312:init] storage-unify: namespace node
> specified as afr-ns
>
> 2008-07-17 15:33:38 D [scheduler.c:45:get_scheduler] scheduler: attempt to
> load file rr.so
>
> 2008-07-17 15:33:38 D [unify.c:4388:init] storage-unify: Child node count
> is 2
>
> 2008-07-17 15:33:38 C [unify.c:4441:init] storage-unify: Initializing
> scheduler failed, Exiting
>
> 2008-07-17 15:33:38 E [xlator.c:261:xlator_init_rec] xlator:
> 'storage-unify' init() failed. Check spec file.
>
> 2008-07-17 15:33:38 E [glusterfs.c:650:main] glusterfs: Error while
> initializing translators. Exiting
>
>
>
> I checked spec file in doc/examples/unify.vol , it is same as version 1.3
>
>
>
> 3.       How to migrate the version from 1.3 to 1.4 ? Do it support on
> line migration?
>
>
>
> Thanks again!  Your help is really appreciated.
>
>
>
> Ben
>  ------------------------------
>
> *From:* anand.avati at gmail.com [mailto:anand.avati at gmail.com] *On Behalf Of
> *Anand Avati
> *Sent:* Thursday, July 17, 2008 1:56 PM
> *To:* Brent A Nelson
> *Cc:* Ben Mok; gluster-devel at nongnu.org
> *Subject:* Re: [Gluster-devel] Performance issue
>
>
>
>
>
> 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...
>
>
> Porting the binary protocol and nbio (which came hand-in-hand into 1.4)
> will be pretty disruptive in the 1.3 series. We are working hard on getting
> 1.4 out soon.
>
>
>
>
> PS The developers are also working towards eliminating the namespace
> requirement for unify/stripe; this could be a significant improvement, as
> well.
>
>
> Currently only unify needs namespace. stripe does not. The new branch will
> have unify as is (with the namespace etc) but a new translator which will be
> a functional replacement to unify will be introduced. This new translator
> (hash) internally builds a distributed hash table over the storage servers
> improving the FS performance significantly. The DHT is similar in concept to
> many of the popular DHT algos, and hashes the filename and stores the file
> on the hashed node. So functionally, it is similar to unify (directory
> structure is replicated, file exists on only one node) but the internals are
> quite different, without a namespace too.
>
> avati
>
> No virus found in this incoming message.
>
> Checked by AVG - http://www.avg.com
>
> Version: 8.0.138 / Virus Database: 270.5.0/1556 - Release Date: 16/7/2008 16:56
>
>


-- 
If I traveled to the end of the rainbow
As Dame Fortune did intend,
Murphy would be there to tell me
The pot's at the other end.



More information about the Gluster-devel mailing list