Details
-
Type:
Sub-task
-
Status:
Open
-
Priority:
Major
-
Resolution: Unresolved
-
Affects Version/s: None
-
Fix Version/s: 3.1.3
-
Component/s: Runtime: Threads and Concurrency
-
Labels:None
-
Number of attachments :
Description
Benchmarks which have many runnables (for example lusearch) are dramatically more stable on systems with more processors. Limiting affinity to a single processor, or running on a machine with fewer CPUs, reduces their stability. It is not entirely clear what this implies, but it probably has something to do with stopping threads for GC. Find the problem and fix it.
Moving all unscheduled issues to 3.1. Please close or retarget to a different fix target as appropriate.