Analysis Report


Tips to reduce GC Time

(CAUTION: Please do thorough testing before implementing out the recommendations. These are generic recommendations & may not be applicable for your application.)

  • 41.06% of GC time (i.e 79 ms) is caused by 'Metadata GC Threshold'. This GC is triggered when metaspace got filled up and JVM wants to create new objects in this space..
    Solution:
    If this GC repeatedly happens, increase the metaspace size in your application with the command line option '-XX:MetaspaceSize'.


JVM Heap Size

Generation Allocated Peak
Young Generation 19.21 gb3.7 gb
Old Generation 12.79 gb15.1 mb
Total 32 gb3.72 gb

Key Performance Indicators

(Important section of the report. To learn more about KPIs, click here)

1 Throughput : 99.475%

2 Latency:

Avg Pause GC Time 15 ms
Max Pause GC Time 60 ms


GC Pause Duration Time Range :

Duration (secs) No. of GCs Percentage
0 - 0.19100.0%

Interactive Graphs

(All graphs are zoomable)










G1 Collection Phases Statistics


initial-mark Young GC Remark Concurrent MarkCleanup Total
Count 4144417
Total GC Time 110 ms40 ms40 ms1 ms0191 ms
Avg GC Time 28 ms40 ms10 ms0011 ms
Avg Time std dev 19 ms000016 ms
Min/Max Time 0 / 60 ms0 / 40 ms0 / 10 ms0 / 00 / 00 / 60 ms
Avg Interval Time 12 sec 23 msn/a12 sec 49 ms12 sec 49 ms12 sec 50 ms12 sec 43 ms

G1 GC Time



Pause Time

Total Time 190 ms
Avg Time 15 ms
Std Dev Time 17 ms
Min Time 0
Max Time 60 ms


Concurrent Time

Total Time 1 ms
Avg Time 0
Std Dev Time 0
Min Time 0
Max Time 0

Object Stats

(These are perfect micro-metrics to include in your performance reports)

Total created bytes 7.35 gb
Total promoted bytes 29.3 mb
Avg creation rate 208.08 mb/sec
Avg promotion rate 829 kb/sec

Memory Leak

No major memory leaks.

(Note: there are 8 flavours of OutOfMemoryErrors. With GC Logs you can diagnose only 5 flavours of them(Java heap space, GC overhead limit exceeded, Requested array size exceeds VM limit, Permgen space, Metaspace). So in other words, your application could be still suffering from memory leaks, but need other tools to diagnose them, not just GC Logs.)


Consecutive Full GC

None.


Long Pause

None.


Safe Point Duration

(To learn more about SafePoint duration, click here)

Total TimeAvg Time% of total duration
Total time for which
app threads were stopped
1.051 secs0.001 secs2.918 %
Time taken to stop app threads0.564 secs0.001 secs1.566 %


GC Causes

(What events caused the GCs, how much time it consumed?)

CauseCountAvg TimeMax TimeTotal TimeTime %
Metadata GC Threshold 326 ms57 ms79 ms41.06%
Others 12n/an/a61 ms31.67%
G1 Evacuation Pause 226 ms41 ms52 ms27.27%
Total 17n/an/a191 ms100.0%

Tenuring Summary

Desired Survivor Size: 1.2 gb,

Max Threshold: 15

AgeSurvival CountAverage size (kb)Average Total 'To' size (kb)
age 1 4 39790.26 39790.26
age 2 3 16002.64 65941.43
age 3 2 20740.56 98678.39
age 4 1 8630.14 157519.28

Command Line Flags

-XX:+AlwaysPreTouch -XX:CompileCommandFile=/etc/dse/cassandra/hotspot_compiler -XX:ErrorFile=/var/lib/cassandra/hs_err_1543414033.log -XX:G1RSetUpdatingPauseTimePercent=5 -XX:GCLogFileSize=10485760 -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=/var/lib/cassandra/java_1543414033.hprof -XX:InitialHeapSize=34359738368 -XX:+ManagementServer -XX:MaxDirectMemorySize=25301090304 -XX:MaxGCPauseMillis=2000 -XX:MaxHeapSize=34359738368 -XX:NumberOfGCLogFiles=10 -XX:+PerfDisableSharedMem -XX:+PreserveFramePointer -XX:PrintFLSStatistics=1 -XX:+PrintGC -XX:+PrintGCApplicationStoppedTime -XX:+PrintGCDateStamps -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintHeapAtGC -XX:+PrintPromotionFailure -XX:+PrintTenuringDistribution -XX:+ResizeTLAB -XX:-RestrictContended -XX:StringTableSize=1000003 -XX:ThreadPriorityPolicy=42 -XX:ThreadStackSize=256 -XX:+UnlockDiagnosticVMOptions -XX:+UseG1GC -XX:+UseGCLogFileRotation -XX:+UseTLAB -XX:+UseThreadPriorities


Mpgyi