Distributed Trunking with HPE 5130?

[Morton28]

I am about to plan a Core Switch for a customer, who has only access switches connected to each other. I'd like to connect every core switch with two fiber connections to each of the two HPE 5130. I read somewhere, that you can only cross-connect if the switches support distributed trunking. The 5130 apparently does not support this.

This seems a little odd and if I had to guess, pretty much every full stack supports this. But before we buy the Core components I'd like to make absolutely sure that we can have the access switches cross connected on both 5130 switches elonone.

Anyone here who can tell me with absolute certainty that this will not be a problem?

Thanks a lot in advance

[parnassus]

Hi @Morton28, I don't see the need to borrow the concept of "Distributed Trunking" (which is a feature of what is known as HP E-Series switch series...feature available on some old HP ProCurve now rebranded Aruba...example: Aruba 5400R zl2) when your HPE 5130, which is Comware OS based, natively supports IRF!

I mean, with two HPE 5130 deployed as an IRF cluster your access switches can easily connect to that IRF core by using normal "Port Trunks" (known also as LAGs Link AGGregations, if they are Comware OS based) using LACP and the IRF stack itself will use standard LAGs with LACP to serve your Access switches.

About the IRF (from HPE FlexNetwork 5130 HI Switch Series' QuickSpecs):

Intelligent Resilient Fabric (IRF) creates virtual resilient switching fabric, where two to nine switches perform as a single L2 switch and L3 router; switches do not have to be co-located and can be part of a disaster-recovery system; servers and/or switches can be attached to IRF using standard LACP Links Aggregations for automatic load balancing and high availability. IRF can eliminate the need for complex protocols like Spanning Tree Protocol, Equal-Cost Multipath (ECMP) or VRRP, thereby simplifying network operation.