For more details, please see ourCookie Policy.

Fibre Channel (SAN)

Occasional Contributor
Posts: 8
Registered: ‎05-06-2009

ISL and virtual channels


I have read several Brocade documents and I´m still searching for some ISL and Virtual channel (VC) best practice/guidance.

For example:
I have a 8510-8 switch with 2 ISL´s in a trunk.
I want to expand the trunk with 2 more ISL´s.

Index Slot Port Address Media  Speed        State    Proto
  16    2    0   50fbc0   id    16G   Online      FC  E-Port  10:00:00:27:f8:30:6f:01 "ssw_1p02_vbas_a02" (downstream)(Trunk master)  
  17    2    1   50fb80   id    16G   Online      FC  E-Port  (Trunk port, master is Slot  2 Port  0 )
  18    2    2   50fb40   --    N16   No_Module   FC  Disabled (Persistent)
  19    2    3   50fb00   --    N16   No_Module   FC  Disabled (Persistent)

How do I balance the trunk using all avaiable VC´s (and buffer credits also)?



Valued Contributor
Posts: 559
Registered: ‎03-20-2011

Re: ISL and virtual channels

you cannot blance anything inside the single trunk. trunking does it for you automatically.

instead of expanding the 2-link trunk to 4 links, i'd advise creating another 2 link trunk using two ports on another port blade. this way you'll have more VCs for data transfer and more resiliency in case if one port blade will fail.
Frequent Contributor
Posts: 141
Registered: ‎05-26-2009

Re: ISL and virtual channels

The VCs are per port and one of the reasons why the next link of the trunk is involved is lack of buffer credits on the first link. So in a normal trunk the traffic on one VC number is not balanced over the ISLs of the trunk. That's not necessary, anyway. There's also no balance between the VCs of a single ISL, because the decision which VC to use is bound to the destination ID. So I'd go with Alexey Stepanov's suggestion to create a second trunk to prevent the fabric from being split if the ASIC fails.

Join the Broadcom Support Community

Get quick and easy access to valuable resources across the Broadcom Community Network.