API Connect V10 -  Running on VMware 

     Hi , 
    
          I have few questions regarding the load balancer in the subsystem , this questions came from the client and it seems the documents only mentioned that it requires a load balancer but did not detailed the exact purposes.

        
         - what is the use of Load Balancer  in APIC subsystem   in a cluster (3 nodes setup ) ,  is it necessary ? 

         -  how does the subsystem cluster replicates / sync the data between each other ? is it via the load balancer ?   

         - if its HA   out of the 3 nodes   who is the primary master , whos is the secondary / slave ?

          -  if one of the node goes down , who will act as the as the master  or what role will be the 2 remaining nodes

           -  if one node goes down ,  once it goes back online does it will sync the datas automatically ?  

            - syncing of the data between nodes will go through the load balancer  ?

Pls kindly advise . Thanks

regards,

Lee

------------------------------
Lee TENEFRANCIA
------------------------------

Look at IBM API Connect v10.x Deployment WhitePaper  for all this information

https://community.ibm.com/community/user/integration/viewdocument/ibm-api-connect-v10x-deployment-wh?CommunityKey=2106cca0-a9f9-45c6-9b28-01a28f4ce947

In brief:

Every type of component needs it's own balancer for incoming requests.

Primary nodes of each type are decided by quorum algorithm automatically. Note (n-1)/2 rule, minimum 3 nodes.

Replication between instances of single type is internal and goes direct between instances (no balancer). Sync mechanism is different for every type of components.

P.S. This is based on v2018 experience, but AFAIK is the same for v10 also.


------------------------------
Denis Migulin
------------------------------