We're looking at some performance issues assocaited with RTC3 over RTC2. They've been dragging along for months now with little direction.

Having dug up the article "https://jazz.net/library/article/557", I've been running over the setup of our various instances.

I'm interested to understand the following settings

Core Repository Component - JDBC Connection Pool Size com.ibm.team.repository.db.jdbc.pool.size max_clients * 25
Core Repository Component - Mediator Pool Size com.ibm.team.repository.rdb.mediator.pool.size max_clients * 25

I received conflicting recommendations regarding the jdbc.pool.size, currently set at default (128), however, we're typically running a peak of around 350 users. We were recommended no to increase this setting, can someone clarify, if we should, and what if any effect there would be if this went upto say 5000. Even that would be short of the 8750 required if we scaled to 350 max_clients.

Also, max_clients, is this users or mean something completely different.

Also, looking at maxThreads for the tomcat server, "Set maxThreads >= (max_clients * 50) + minSpareThreads for enabled Connectors", again, 350 concurrent clients would require 17500+25 to meet this need, currently set at 13550.

I/m not a tomcat tuning expert so any advice or implications would be gratefully received.