Active/Active HA3 Link Bandwith

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Announcements

Active/Active HA3 Link Bandwith

L1 Bithead

Looking for advice on bandwidth for a A/A pair of 5050s running a 10G vwire. Should I burn another 10G port on the HA3 link or can I get away with 3 to 5G agg? Or is this really just one of those things I need to set up and monitor?

1 accepted solution

Accepted Solutions

L5 Sessionator

If you just have on 1 ingress(10G) and egress (10G ) you should be okay with 3 to 5G aggregate . But I would still recommend you to setup and monitor.

View solution in original post

4 REPLIES 4

L5 Sessionator

If you just have on 1 ingress(10G) and egress (10G ) you should be okay with 3 to 5G aggregate . But I would still recommend you to setup and monitor.

L5 Sessionator

sraghunandan is right, two thin takes in consideration:

     HA3 bandwidth: depend of you load balancing method and your election (setup owner / Setup setup) method => need to mpnitor

     HA3 backup: if your HA3 link is down your cluster ... is not healthy. The only way for HA3 ha is aggregate.

Then in my ming: better to have 3 to 5G with agg then only one 10gG.

v.

According to the HA-active-active-technote:

"

The HA3 link is used for packet forwarding between the session owner and the session setup device in an active-active cluster. HA3 link is a layer2 link and uses MAC-in-MAC encapsulation. Aggregate interfaces can be configured as a HA3 link on the PA-5000 and PA-4000 Series. This also provides redundancy of HA3 link. The interface that will be used as HA3 link must be set as type HA

Note: No backup links can be configured for the HA3 link. Use the aggregate interface on the PA-5000 and PA-4000 series if redundancy is required.

"

This gives if you use multiple 1G interfaces as aggregation for HA3 in an active-active setup where the up/downlinks are 10G then each session through the PA-cluster might be limited to max 1Gbit/s depending on if HA3 is used in the flow or not. In this case you could check if you can setup the two PA devices as independent VWIRE instead and let the router/switch before/after the PA-cluster run portchannel and "loadbalance" the sessions so its loadbalanced on dstip one way and srcip the other way.

I'm thinking about setting up 4-1G port as my aggregate setup to my edge switch. My issue is understanding layer 2 adjacent which needs to be setup on the switch and what's going across my link to the other data center, what triggers the sync, and how much data is going across this link?

  • 1 accepted solution
  • 4834 Views
  • 4 replies
  • 0 Likes
Like what you see?

Show your appreciation!

Click Like if a post is helpful to you or if you just want to show your support.

Click Accept as Solution to acknowledge that the answer to your question has been provided.

The button appears next to the replies on topics you’ve started. The member who gave the solution and all future visitors to this topic will appreciate it!

These simple actions take just seconds of your time, but go a long way in showing appreciation for community members and the LIVEcommunity as a whole!

The LIVEcommunity thanks you for your participation!