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

how to configuring a cluster farm

m_larocca
Newcomer

Hi community,

i have a lot of doubts about how to start a cluster, i have read the documentation but i want more information about your expirience.

You can se my actual configuration in the follow image:

And i'd like that my environment will be like the follow image:

I explain what i want do and what are my goals.

As you can see i want associate the AMD CED FS on both CAS because i want balance the number of SS from the AMD ced FS on the two CAS.

So i want configure the first Cluster with:

--CLUSTER A--

CAS ( own SQL 1 )

AMD ced fs

---------------------

--CLUSTER B--

CAS 2 ( own SQL 2 )

AMD ced fs

AMD SALUTE

AMD AGEA

---------------------

For example now my AMD CED FS has 100 software service when i will do the farm like in the image up i wish i have balanced 50 SS on CAS and 50 SS on CAS 2. ( it is possible ? )

Consider that every AMD has a unique customer(FS,SALUTE,AGEA), the primary BIG customer traffic come from the ADM CED FS that has 100 SS, the others 2 AMD traffic is minor, they has 20 SS SALUTE and 20 SS AGEA.

After this i want that all application users log in only in the primary CAS of the cluster A, a part of this users have to see only the SS configured on AMD SALUTE( that is configured in cluster B ), other part of the users have to see only the SS configured on AMD AGEA( that is configured in cluster B). It is possible ?

If i configure the farm like i described can i roolback ?

Thanks for the answer.

Marco

1 REPLY 1

It is possible to achieve what you described.
What you need to configure is:

  1. create a farm as described
  2. define software service filters on both clusters: on cluster A list all 50 software services you want to process there and on cluster B list all SS you want to process from ALL probes
  3. to restrict the access to certain software services for specific users use data access permission mechanism

If you perform these steps you can rollback but you would not be able to change the historical data (Cluster A would have data for 50 software services for the time the farm existed)