Another test of 8MB regions, this time with -XX:G1RSetUpdatingPauseTimePercent=5
.
Baseline
|
8GB heap, G1HeapRegionSize of 8MB, RSetUpdatingPauseTimePecent of 5
|
---|---|
Cassandra read rate | |
Cassandra write rate | |
Cassandra read latency (75p) | |
Cassandra read latency (99p) | |
Cassandra write latency (75p) | |
Cassandra write latency (99p) | |
Cassandra connection timeouts/sec | |
Collection time (accumulated, 5 minute moving average) | |
Collections | |
Only one allocation failure here, resulting in a young collection being extended to ~1.9s (selected out of scatter plot above).
17 humongous allocations resulted in premature starts to concurrent cycles (a significant improvement over observations at 4MB region size).