Re: Help ZFS FreeBSD 8.0 RC2 Write performance issue

On Thu, Nov 12, 2009 at 11:02 AM, Rick Macklem <rmacklem_at_uoguelph.ca> wrote:
>
>
> On Wed, 11 Nov 2009, Ivan Voras wrote:
>
>>
>> I think NFS uses sync disk IO access by default, this may be your
>> problem if you are write-heavy. Try setting vfs.nfsrv.async to 1 to
>> see if this is the cause of your problems.
>
> Just fyi, I took a quick look and I don't think this will be a good
> idea for NFSv3. (It allows the server to cheat for NFSv2 and avoid
> synchronous writes, which was contrary to the standard, but became
> fashionable for performance reasons, before NFSv3 came out.)
>
> For NFSv3, the writes are normally done asynchronously, followed
> by a Commit RPC to force them to disk, done by the client when it
> is flushing its buffer cache.
>
> When you set this sysctl, the NFSv3 server (sys/nfsserver, not the
> experimental one), all it does is reply to the write RPC that it
> has already been committed. This may have two effects, depending
> upon the client:
> 1 - The client may then choose to not bother with a Commit RPC.
>    --> This shouldn't help performance much, because the data
>        will normally have made to disk by now.
> *** This might be an interesting experiment to try on a ZFS server
>    though, since if it does make a significant difference, it
>    suggests that ZFS does a lot of work figuring out that the
>    blocks are already on stable storage or something like that.
>    (ie. It might hint at where to look for a ZFS related perf.
>     problem.)
> OR
> 2 - Nothing, because the client doesn't notice it doesn't need to
>    commit it and does the Commit RPC anyhow.
>
> Also, it is potentially dangerous, since if the server crashes after
> the client has done the write, but before the server has written it
> to disk, the data may be lost. (ie. The client might have flushed the
> dirty blocks out of its buffer cache, because it didn't think it needed
> to do a Commit RPC.)
>
> rick


I guess I am confused a bit.
Why is it that sftp has better write speed than NFS? and from what I
am hearing there isn't a good way to fox it?
what about if I try and tweak the nfs connection paramaters. all of
these tests were done with whatever default is.

anyone have any Ideas on better client connection parameters?


Sam Fourman Jr.