Hello,
I adapted the JVM settings as follows:
-server -XX:PermSize=256m -Xms1g -Xmx8g -XX:+UseParNewGC
-XX:+UseConcMarkSweepGC -XX:+UseTLAB -XX:NewSize=128m -XX:MaxNewSize=128m
-XX:MaxTenuringThreshold=0 -XX:SurvivorRatio=1024
-XX:+UseCMSInitiatingOccupancyOnly -XX:CMSInitiatingOccupancyFraction=60
-XX:+DisableExplicitGC
The first SQL now runs as follows:
executeQuery: 1017ms
rs.next: 15118ms
> Which is 5 seconds faster on the query result iteration.
The big sql still runs very long (but without errors).
Based on the log I don't think it is in need of memory now and GC also looks
under control:
/Metrics for local node (to disable set 'metricsLogFrequency' to 0)
^-- Node [id=f4f5b285, name=null, uptime=00:28:00:175]
^-- H/N/C [hosts=1, nodes=1, CPUs=8]
^-- CPU [cur=13.73%, avg=17.07%, GC=0.17%]
^-- Heap [used=2890MB, free=64.72%, comm=4805MB]
^-- Non heap [used=263MB, free=13.34%, comm=273MB]
^-- Public thread pool [active=1, idle=15, qSize=0]
^-- System thread pool [active=0, idle=16, qSize=0]
^-- Outbound messages queue [size=0]
INFO [20160504 08:51:56] -
Metrics for local node (to disable set 'metricsLogFrequency' to 0)
^-- Node [id=f4f5b285, name=null, uptime=00:29:00:192]
^-- H/N/C [hosts=1, nodes=1, CPUs=8]
^-- CPU [cur=13.7%, avg=16.97%, GC=0.2%]
^-- Heap [used=2800MB, free=65.81%, comm=4805MB]
^-- Non heap [used=263MB, free=13.34%, comm=273MB]
^-- Public thread pool [active=1, idle=15, qSize=0]
^-- System thread pool [active=0, idle=16, qSize=0]
^-- Outbound messages queue [size=0]/
Minutes further:
Metrics for local node (to disable set 'metricsLogFrequency' to 0)
^-- Node [id=c8a9b84a, name=null, uptime=00:28:00:136]
^-- H/N/C [hosts=1, nodes=1, CPUs=8]
^-- CPU [cur=13.53%, avg=15.33%, GC=0.2%]
^-- Heap [used=2671MB, free=67.39%, comm=4691MB]
^-- Non heap [used=156MB, free=48.68%, comm=266MB]
^-- Public thread pool [active=1, idle=15, qSize=0]
^-- System thread pool [active=0, idle=16, qSize=0]
^-- Outbound messages queue [size=0]
At this point I just kill the server ;-).
br
jan
--
View this message in context:
http://apache-ignite-users.70518.x6.nabble.com/Unexpected-performance-issue-with-SQL-query-followed-by-error-tp4726p4750.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.
