[f-nsp] Loadbalacing (Trunking/LACP) on foundry MLX: unequal load distribution.

[William A. Cooper]

Hi,

IEEE  802.3ad does not define a load distribution algorithm (in fact 
purposefully states
it is vendor specific).

Trunk Server typically denotes L3 based load distribution, while trunk 
switch denotes L2; however
I defer to more experienced Foundry folk, and would suggest contacting 
your local Foundry rep to
help you out-

-Tony

Rolf Hanßen wrote:
> Hi,
>
> "trunk server" works fine on the xmr we use.
>
> trunk server ethe 1/19 to 1/20
>
> SSH at bbr1-dus1#show interfaces e 1/19 to 1/20
>   300 second output rate: 531005356 bits/sec, 64551 packets/sec, 53.72%
> utilization
> ...
>   300 second output rate: 488615879 bits/sec, 61202 packets/sec, 49.44%
> utilization
>
> But we are not using LACP, don't know if that has any influences here.
>
> Regards
> Rolf
>
>   
>> On Monday 05 February 2007 20:00, Gunther Stammwitz wrote:
>>     
>>> Any idea how we could optimize this? How does the
>>> loadbalancing algorith on the mlx work?
>>>       
>> Looks like the balancing algorithm is similar for XMR and MLX:
>>
>> http://www.foundrynet.com/services/documentation/xmr_user/Trunking.html
>>
>> --
>> Gerald    (ax/tc)
>> _______________________________________________
>> foundry-nsp mailing list
>> foundry-nsp at puck.nether.net
>> http://puck.nether.net/mailman/listinfo/foundry-nsp
>>
>>     
>
>
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
>
>