Question Should we use dedicated JVMs as HA coordinators. And if so are the recommendations different between WAS versions 7.0, 8.0 and 8.5?
Yes.
Recommendations are same across v7 and v85
It is a best practice to configure one or more preferred coordinator processes for each core group. This limits the movement of the coordinator and number of state rebuilds.
Configuring preferred coordinators is a best practice.
Ideally, assign processes that do not host applications and are located on machines with spare capacity as preferred coordinators.
http://www3.software.ibm.com/ibmdl/pub/software/dw/wes/0710_largetopologies/LargeWebSphereTopologies.pdf
Question Should we enable HA for applications that utilize ejb, XA, Websphere MQ, or sibbus? And if so are the recommendations different between WAS versions 7.0, 8.0 and 8.5? What are the benefits of enabling it? Are there any concerns?
The components that use HA services in WAS are
- EJB workload management
- JMS workload management
- On-demand configuration routing
- Data replication service
- HTTP Session distribution when using memory to memory replication
- Dynamic cache distribution, when using memory to memory replication
- Stateful session bean failover, which relies on memory to memory replication
- Session Initiation Protocol (SIP)
If you need any of these workload management, replication and routing services then you need to enable HAM on the cluster. The downside to enabling HAM is increased CPU activity across all the cluster JVMs and increase in network chatter due to data replication across JVMs. Logs get filled with HAM membership messages as servers come up and down. Recommendations are same across v7 and v8.5
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.