Analyzing Bottlenecks on the Grid Engine Master

To provide effective performance tuning, you must understand the bottlenecks of distributed systems. The $SGE_ROOT/dtrace/monitor.sh script measures throughput-relevant data of the running Grid Engine master and compiles this data into a few indices that are printed in a single-line view per interval. This view shows four main categories of information:

• Spooling — Indicates the number of operations that spooled to the qmaster process and the elapsed time

• Request handling — Shows the number of messages sent and received of various types, such as reports, GDI requests, and ACK messages.

• Scheduling — Indicates the number of scheduling requests sent to the schedd process and the elapsed time

• Qmaster processing — Includes information about qmaster/schedd communications, qmaster requesst I/O activities, and qmaster lock and unlock requests.

For more information, see the example below.