IBM Tivoli Monitoring > Version 6.3 Fix Pack 2 > Installation Guides > Installation Guide > Performance tuning > Agentless Monitoring
IBM Tivoli Monitoring, Version 6.3 Fix Pack 2
To monitor many monitoring servers, spread the target monitoring servers across multiple process instances
Although it is possible to monitor 100 target monitoring servers from a single process instance, if historical data collection is used the attribute group data from all of the target monitoring servers is written to a single attribute group file.
This exhibits the same behavior as keeping agent historical data on the monitoring server rather than the monitoring agent. When a Tivoli Enterprise Portal client requests short-term historical data from a single target system, the entire attribute group file must be ready to satisfy the request.
To monitor a large number of target monitoring servers using agentless monitoring, it is better to use multiple process instances on the same system, and spread out the target monitoring servers across the process instances. This keeps the number of target monitoring servers with data in a single file to a smaller number, and improves response time for Tivoli Enterprise Portal requests for target system short-term historical data.
Parent topic:
Agentless Monitoring