Testing the GNIC-II (Hamachi) gigabit NIC

Leonardo Valcamonici leonardo.valcamonici@caspur.it
Fri Nov 6 08:56:48 1998


On 6 Nov 1998, Jes Sorensen wrote:

> >>>>> "Alan" == Alan Cox <alan@lxorguk.ukuu.org.uk> writes:
> 
> >> to recognize that there is a 64-bit card in a 32-bit PCI slot), I'm
> >> trying to do some benchmarks. The cards are in a back-to-back
> >> configuration between an Alpha-533 and a PentiumII-266.
> >> 
> >> Are there any TCP settings I can possibly tweak to improve TCP
> >> performance on large file transfers ?
> 
> Alan> The biggest thing you want to do is to use 8K packets if the
> Alan> card can support it, instead of 1500 byte frames for gig-E as
> Alan> 'aggregation media'
> 
> Right, for me with the Alteon card, the performance increases from
> like 28MB/sec to around 90MB/sec in a P2 400MHz 440BX (PCI 32/33)box
> when going from 1500 byte frames to 9K ones.
> 
> However, I would be interested to hear what kind of performance you
> get on 1500 byte frames.
> 

Talking about Hamachi cards and not Alteon's, is there anyone out there
who has ever succeeded in increasing MTU in the Becker's 0.07
hamachi driver ? I can only see 21MBps on TCP for back to back connections
between two PentiumPro 200 Linux 2.'.34 boxes...

Cheers,

Leonardo

+==========================================================================+
| Leonardo Valcamonici                                                     |
| CASPUR                                     e-mail: valcamonici@caspur.it |
| Italian Interunivesities Consortium        voice : +39-6-49913857        |
| for Supercomputing Applications            fax   : +39-6-4957083         |
+==========================================================================+
|  The most exciting phrase to hear in science, the one that heralds new   |
|    discoveries, is not "Eureka!" (I found it!) but "That's funny ..."    |
|                             -- Isaac Asimov                              |
+==========================================================================+

 | To unsubscribe, send mail to Majordomo@cesdis.gsfc.nasa.gov, and within the
 |  body of the mail, include only the text:
 |   unsubscribe this-list-name youraddress@wherever.org
 | You will be unsubscribed as speedily as possible.