Re: [Gluster-devel] Multiple NFS Servers (Gluster NFS in 3.x, unfsd, knfsd, etc.)

[Gordan Bobic]

Anand Avati wrote:
> > So - I did a redneck test instead - dd 64MB of /dev/zero to a file on the
> > mounted partition.
> >
> > On writes, NFS gets 4.4MB/s, GlusterFS (server side AFR) gets 4.6MB/s.
> > Pretty even.
> > On reads GlusterFS gets 117MB/s, NFS gets 119MB/s (on the first read after
> > flushing the caches, after that it goes up to 600MB/s). The difference in
> > the unbuffered readings seems to be in the sane ball park and the difference
> > on the reads is roughly what I'd expect considering NFS is running UDP and
> > GLFS is running TCP.
> >
> > So in conclusion - there is no performance difference between them worth
> > speaking of. So what is the point in implementing a user-space NFS handler
> > in glusterfsd when unfsd seems to do the job as well as glusterfsd could
> > reasonably hope to?
>
> Can you clarify if your tests had a setup where the NFS re-export
> would result in 2 hops for IO? From what you have shown, it looks like
> both the tests had just one (physical) network hop (not considering
> loopback).

There are 3 servers in AFR, so I would have thought the write test would 
require the write to get propagated to all the servers, which means the 
primary server the client is connecting to, plus the two other servers. 
Or does that not could as multiple hops in terms of what you are describing?

> The need for an NFS xlators becomes apparent when you want
> to re-export a distributed glusterfs configuration via NFS which can
> result in > 1 network hops for IO. Context switches between these two
> hops makes things considerably worse, and having NFS xlator inside
> glusterfs makes it use the caches in the performance translators very
> effectively.

NFS has FS-Cache support on the client. I would have thought this would 
address that issue.

Gordan