Network Management

 View Only
  • 1.  Link Aggregation Configuration to achive 20Gbps

    Posted Dec 19, 2023 12:06 PM

    Hi, I want to bond two pairs of 10Gbps ports to create 20Gbps bandwidth. So I have two 10G ports connected to my NAS and another two 10G ports connected to my server. I created two trunks, one in static mode for the server and another in LACP mode for the NAS. But still, in the Port Configuration interface, it shows only 10Gbps on the trunk connection. What do I do wrong, or is it even possible to create such a bond?



  • 2.  RE: Link Aggregation Configuration to achive 20Gbps

    Posted Dec 19, 2023 01:11 PM
    Hi! Is it maybe a "visualization" bug of the Aruba Instant On switch you're using not reporting the aggregated speed? how respectively your NAS and your Server are seeing their links aggregation? are they Up (I think so)? do they show 20Gbps aggregated or what?

    You should ask help on the Aruba Instant On community here:






  • 3.  RE: Link Aggregation Configuration to achive 20Gbps

    Posted Dec 20, 2023 03:08 AM

    Yeah, both server and NAS show 20Gb only aruba shows 10Gbps and when I do some test like copying files from nas to server it also shows 10Gbps




  • 4.  RE: Link Aggregation Configuration to achive 20Gbps

    Posted Dec 20, 2023 03:21 AM

    Using one source and one destination will just use one link in the LAG.

    Some information about LAG: https://en.wikipedia.org/wiki/Link_aggregation



    ------------------------------
    Arne Opdal
    ------------------------------



  • 5.  RE: Link Aggregation Configuration to achive 20Gbps

    Posted Dec 20, 2023 05:26 AM

    Exactly! I would not use the "it goes only up to 10Gbps" argument against bonding and LACP not using the aggregated bandwidth (10+10=20Gbps)...the point is to understand how bonding (links aggregation) works...and not pretending to reach 20Gbps when each transmission flow uses a particular LAG member interface. In a scenario where the egress traffic is going to be generated in a "many-sources-to-many-targets" (and the load balancing leverages that) probably both interfaces are fully used and the aggregated bandwidth available is all consumed (note: aggregated).