Details
-
Type:
Improvement
-
Status: Open
-
Priority:
Major
-
Resolution: Unresolved
-
Affects Version/s: 0.2.0
-
Fix Version/s: None
-
Component/s: Server
-
Labels:None
Description
From running the cluster perf run, I see these two locks as being contended.
Metric can probably be turned to atomic instructions and we're doing something expensive with the query exec state lock.
ImpalaServer::query_exec_state_map(52057): 1m3s Metric(1528664): 3s729ms