[vortex] Performance tests with 3c905B

[Thomas Ackermann]

Hi!

I tried to tune my network and did some tests with two Pentium II
hosts at 350 and 400 MHz, both equiped with a 3c905B at 100 Mb

I tested with 3 files (1MB, 10MB and 100MB) over "rsh" and got this
results (in kB per second) for ten runs each:

1MB File:
2245.61403509
5361.2565445
5389.47368421
5361.2565445
5361.2565445
5389.47368421
5361.2565445
5361.2565445
5305.69948187
5361.2565445

10MB File:
2902.49433107
10301.8108652
10291.4572864
10281.124498
10168.8182721
10189.0547264
10158.7301587
10301.8108652
5838.08437856
10291.4572864

100MB File:
6257.63871914
5899.29715405
5970.14925373
5140.0461801
6266.44636191
6209.44757747   
6245.80664837
6236.67702053
6102.13932424
9186.32816004

I tested with that commandm reading file from the filesystem and coping to
/dev/null on the destination host:

(for f in test1 test10 test100 ; do for i in 1 2 3 4 5 6 7 8 9 10 ; do
echo "File: $f" ; time rcp $f amerika:/dev/null ; sleep 10 ; done ;echo
"#####" ;  done ) 2>&1 | tee test.log

So, 5 to 6 MB would be what i expected, but 9 and even over 10 MB are *not*
what i expcected ...

So, my questions is, did i test it wrongly or what shall i think about
my performance test?

***

By the way, what do you think or know, is the *best* 100Mb PCI-NIC for
Linux - regarding sheer performce (or maybe what's the best such NIC for
a server and for a client)?

	Thanx for any input and
	Byebye,
-- 
        Thomas Ackermann | Tel. +49-(0)228/631369 | Mobil: 0178-2016033
                finger tgm@rhein.math.uni-bonn.de for public key
                         Email: <tgm@math.uni-bonn.de>