lkml.org 
[lkml]   [1998]   [Dec]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: NFS ...
Pavel Machek wrote:
>
> Hi!
>
> > * Performance could be better. A few hundred kB/s should be possible on a
> > not to busy 10MBit network, should it? I often only saw 40kB/s, but the
> > server machine was slow (486-100), too. Now with P-150, I see 100kB/s
> > between two Linux systems, which is still a little low, isn't it?
>
> Performance definitely should be _much_ better. 486DX4/100 is fast
> enough machine for such tricks. Link is 100Mbit ethernet, everything
> is loaded.
>
> Tests are done with various big files (>1.5Meg). Speeds are in K/sec
>
> Server 2.0.X, client 2.0.X
>
> FTP:
> server pentium/100, client pentium/100: 3700 (cache-cold), 5400
> (cache-hot), 4100 (cache-cold), 4400 (cache-hot)
>
> NFS:
> server pentium/100, client pentium/100: 1186 [bogus because already in
> clients cache?], 320 (cache-cold)
>
> Oops, oops, nfs performance is really bad.
>

Well, laptop P166MMX with 3Com PCMCIA client to my desktop AMD K6-200
with a cheap ne2k-pci on a RG58 cable copies the binary tarball of my
test Oracle installation in

[root@dilbert /mnt]# time cp 805bin.tar.gz /tmp
0.08user 14.83system 1:54.38elapsed 13%CPU (0avgtext+0avgdata 0maxresident)k
0inputs+0outputs (80major+10minor)pagefaults 0swaps

so it means

[asuardi@dilbert /tmp]$ ls -l 805bin.tar.gz
-rw-r--r-- 1 root root 73484970 Dec 20 16:50 805bin.tar.gz
[asuardi@dilbert /tmp]$ bc -ql
73484970/114.38
642463.45514950166112956810

A second attempt finishes in 114.91 seconds. 627KB/s looks nice on a
cheap home network :) of course there is only me ;)


Client + server are RH5.2 systems with HJ's knfsd-981204.

--alessandro <asuardi@uninetcom.it> <asuardi@it.oracle.com>

Linux 2.0.36/2.1.131 glibc-2.0.7-29 gcc-2.8.1 binutils-2.9.1.0.15

"I hate bugs which disappear just as soon as you start trying to
narrow things down." -- Stephen Tweedie



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:46    [from the cache]
©2003-2011 Jasper Spaans