Tuning heartbeat and cluster parameters

We can work on both WebSphere connection techniques and tuning HACMP or HACMP/ES parameters to enhance system performance. WebSphere V5 has significantly improved failure-handling techniques compared to earlier versions of WebSphere. HACMP 4.5 provides easier and greater control over several tuning parameters that affect the cluster's performance. Setting these tuning parameters correctly ensures throughput and adjusting the HACMP failure detection rate can help avoid failures caused by heavy network traffic.

Some options are: Improving WebSphere connection pool techniques to minimize the failures when reconnecting after a failover. We fully tested WebSphere V5.x, WebSphere V4.x, WebSphere V3.5.x, and WebSphere V3.02 with HACMP 4.5, HACMP 4.3.1, HACMP 4.4, HACMP 4.4.1, and HACMP/ES for both DB2 and Oracle. The entire end-to-end high availability system performed very well. WebSphere V5 and WebSphere V3.5.3 and later versions have significantly improved connection pool techniques to minimize the failures during an HACMP failover, compared to WebSphere V3.02 and WebSphere V3.5.2 tested with HACMP 4.3.1. Adjusting high and low watermarks for I/O pacing. The default value is 33 -24. Adjusting syncd frequency rate. The default value is 10. Adjusting HACMP failure detection rate. There are two parameters that control the HACMP failure detection rate are: HACMP cycles to failure: the number of heartbeats that must be missed before detecting a failure. HACMP heartbeat rate: the number of microseconds between heartbeats.

For example, if heartbeat rate=1 second and cycles=10, the failure detection rate would be 10 seconds. Faster heartbeat rates may lead to false failure detection, particularly on busy networks. Please note that the new values will become active the next time the cluster services are started. AIX deadman switch timeout. If HACMP for AIX cannot get enough CPU resource to send heartbeats on IP and serial networks, other nodes in the cluster will assume the node has failed, and initiate takeover of the node's resources. In order to ensure a lean takeover, the deadman switch crashes the busy node if it is not reset within a given time period.

  Prev | Home | Next

 

WebSphere is a trademark of the IBM Corporation in the United States, other countries, or both.

 

IBM is a trademark of the IBM Corporation in the United States, other countries, or both.