5

I'm using GlusterFS for a file storage cluster and my web apps.

I have two volumes, the storage and the web ones.

Currently I have the web one mounted as NFS, because FUSE was way too slow.

Now I was wondering if there's any big disadvantage and if I should mount the storage one as NFS too.

2

NFS uses the standard filesystem caching, the Native GlusterFS uses up application space RAM and is a hard-set number that must defined.

source

If you look at the documentation

The FUSE client allows the mount to happen with a GlusterFS round robin style connection. In /etc/fstab, the name of one node is used; however, internal mechanisms allow that node to fail, and the clients will roll over to other connected nodes in the trusted storage pool. The performance is slightly slower than the NFS method based on tests, but not drastically so. The gain is automatic HA client failover, which is typically worth the effect on performance.

So in summary: It depends on the requirements. When there is enough RAM available on the system and performance is not overall important FUSE seems to be the better choice.

2
  • 1
    From my experience, the performance differences are huge. After switching my web app from FUSE to NFS load times decreased from 1.5 - 4 seconds to under 1 second. Also I tried extracting some archives today and it seems to take 4-5 times longer on FUSE. The failover is great, but I also need the performance. Is there anything I can do about that? – Alex Dumitru Aug 10 '16 at 19:13
  • I think so. But it would need some effort access.redhat.com/documentation/en-US/Red_Hat_Storage/2.0/html/… – Henrik Pingel Aug 10 '16 at 19:22
2

I have one FTP mount under one native GlusterFS. I sent many files to this FTP and the performance was terrible. When I changed to an NFS mount the performance was much better than Native GlusterFS, in my case of course.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.