best performance options for nfs

[ESGLinux]

2010/4/22 Romeo Theriault <romeotheriault at gmail.com>

> >
> > So I want to improve the performance of this nfsserver.
> >
> > Do you know which options are best for this? any advice ? (I have read
> that
> > using nscd is good for it)
> >
>
>
> If your switch supports it you'll probably want to enable jumbo frames,
> mtu,
> on the server, switch and the clients. It's probably also a good idea to
> set
> your rsize ,wsize mount options to 32768. The wsize and rsize options
> specify the size of the chunks of data that the client and server pass back
> and forth to each other. Obviously, the more data you can transfer in one
> go
> the better. Depending on the needs of the applications you might get a bit
> of a performance boost by disabling access time updates on the nfs server.
>
> Another performance booster for writes at the expense of possible data
> corruption if a reboot were to occur during writes is the use of the async
> option. This allows the server to tell the client everything was
> successfully written before it actually all has.
>
> Check out this link:
>
>
> http://www.cyberciti.biz/faq/linux-unix-tuning-nfs-server-client-performance/
> *
> *
>
>
I have read your post and the link, and now I have doubt:

Note: Setting this size to a value less than the largest supported block
size will adversely affect performance.

How can I check I set the optimus value (or at least not a bad value)

Greetings

ESG


> --
> Romeo Theriault
> System Administrator
> Information Technology Services
> --
> redhat-list mailing list
> unsubscribe mailto:redhat-list-request at redhat.com?subject=unsubscribe
> https://www.redhat.com/mailman/listinfo/redhat-list
>