On 5/12/06, Gus Wirth <[EMAIL PROTECTED]> wrote:
If I do an NFS mount and copy a large file from the server to /dev/null on the client or if I scp a file between the server and client I get pretty much the same results, about 11MB (that's bytes) per second. This is only about 10% better than 100baseT ethernet.
NFS (unless forced to tcp) copy is not a good test to test throughput. With a crappy switch, udp nfs will have a lot of retries. Copy the file and check the nfs-client stats. Also, the default read and write window sizes are small in NFS. Force it to 32k. (the man pages should show you how to do both.) Are you using atleast cat5e or cat6? Are the connections really gigE ?. Check the boot log. Maybe you can force gigE using mii-tool. -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
