One of the most common problems in administering a Sybase Replication Server is measuring performance and identifying the possible causes of performance issues when they occur. Interestingly enough, in both cases, most Replication Server administrators use the same two techniques that can at best be considered symptoms:
1. Queue disk space consumption for a connection
2. Latency between the primary and replicate — either via attempts to use rs_lastcommit or a separate timing table that is periodically updated
The problem is that while both can be used to identify when there is or isn’t a performance issue, neither is very useful beyond simply raising an alert to a possible problem. Unfortunately, the ...