In multithreaded programs, synchronization wait tracing keeps track of real-time delays in calls to thread-synchronization routines if the delay exceeds a predetermined threshold.
Synchronization tracing for applications written in the Java TM programming language is based on events generated when a thread attempts to acquire a JavaTM monitor. Both machine call stacks and Java call stacks are collected for these events, but no synchronization tracing data is collected for internal locks used within the JavaTM Virtual Machine (JVM) software. For such programs, synchronization tracing data collection uses JavaTM Virtual Machine Profiler Interface (JVMPI) events, which causes dilation in proportion to the amount of monitor contention in the application.
Note: the terms "Java Virtual Machine" or "JVM" mean a virtual machine for the Java platform.
Synchronization wait tracing information is converted into the following metrics:
Metric | Definition |
---|---|
Synchronization delay event count | The number of events in which the delay exceeds the threshold |
Synchronization wait time | The time spent waiting on calls to synchronization routines. Only the events for which the wait time is greater than the threshold are included in this sum. |
See also | |
---|---|
Timing Metrics Hardware Counter Metrics Memory Allocation Metrics MPI Tracing Metrics Exclusive, Inclusive, and Attributed Metrics Setting Data Presentation Preferences |