ServerNet Cluster Manual
Troubleshooting and Replacement Procedures
ServerNet Cluster Manual—520575-003
7-39
Diagnosing Performance Problems
Diagnosing Performance Problems
Diagnosis of performance problems in any environment involves multiple steps and 
requires extensive knowledge of performance fundamentals and methodologies. If 
there are ServerNet Cluster performance issues, you might want to explore these 
areas:
•
Throughput: How much traffic is flowing through nodes and between individual 
processors?
•
Latency: How well does traffic flow?
•
Problem isolation: What is responsible for the problem?
The Measure performance tool can assist you in collecting data for throughput and 
latency analysis. Measure has a SERVERNET entity that tracks all the interprocessor 
communication for remote nodes (remote IPC type). This information allows you to 
isolate and examine the flow of ServerNet cluster traffic. Another area to investigate, if 
you suspect performance degradation, is the health of your EXPAND lines. For more 
information on the SERVERNET, NETLINE, and PROCESS entities and the G08 
counters added specifically for ServerNet cluster support, consult the Measure 
Reference Manual. 










