lkml.org 
[lkml]   [1999]   [Mar]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectPoor NFS performance?
Hi all,

today I tested the harddisk-performance of my server (EIDE/2.0.36),

dd if=100M-File of=/dev/null needed 16.5sec ==> 6M/sec (good for an
old AMD K5 based on an old mainboard).
This server is direct connected to my other mashine via 5m C5 X-link
cable (this should be ok).
Both mashines have a "D-Link somewhat"-card based on DEC 21140
("tulip"/100MBit).

Reading the same file over nfs on my other mashine´s /dev/null needs
>50 seconds, this is less than 2 M/sec.

What is the problem? Where is the real "bottleneck" (is this right
word in english too?) on the data´s way?

technical details of the mashines:

server (rhb): "other" (rhb1):
CPU: AMD K5@133 MHz AMD K6-2 300 @300 MHz
RAM: 64M EDO 96M SDRAM
NIC: ------- D-Link based on DEC 21140/ 100MBit --------
HDD: IBM DTTA#??? (10 GB, 5400 RPM) isn´t interesting

SYS: 2.0.36 (no patches) 2.2.1 (no network-relevant patches)

Have a look on the bing-results in the attachment.

What else do you need? What else may be useful to find the real
bottleneck?

Greetings
Raphael Becker
--
_ _ Powered by SuSE___ ___ _
| | (_)_ _ _ ___ __ |_ ) |_ ) / |
| |__| | ' \ || \ \ / / / _ / / _| |
|____|_|_||_\_,_/_\_\ /___(_)___(_)_|[unhandled content-type:application/octet-stream]
\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.031 / U:0.880 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site