Tideway provides patterns for the following industry leading cluster and high availability products which enable customers to:
- Verify resiliency is built into the infrastructure running their critical business applications
- Verify and compare configuration of cluster nodes

Click on the links to learn more about these patterns:
- HP ServiceGuard: http://www.tideway.com/tknwiki/index.php/HP_ServiceGuard
- IBM HACMP (High Availability Cluster Multi Processing):
- Sun Cluster: http://www.tideway.com/tknwiki/index.php/Sun_Cluster
- Veritas Cluster Server:
- Microsoft Cluster Server: http://www.tideway.com/tknwiki/index.php/Microsoft_Cluster_Server

We are very interested to learn:
- Which additional cluster and high availability products you would like Tideway to identify?
- The specific configuration issues you face when managing clusters and what configuration data you need to help manage these?

There’s other value here - this type of technology usually makes it into engineering standards. I have seen that such standards are often not well used in practice:

- Core O/S capabilities meet or exceed the needs of niche HA / Clustering products
- HA/Clustering is tough and really needs to be in the transaction layer (which is usually, but not always, in the dbms). Because it’s so tough, I’ve seen BCP approaches that rely on manual failover, as the app owners don’t trust the testing of the automated failover in the cluster to be reliable.
- the resilience of the dbms products has improved dramatically in recent years: and that’s all that’s needed for most business apps.

Unless the engineering standards are reviewed against actual take up, there’s likely to be significant overspend.