JVM Instances |
jvm_Ctr_1(1), jvm_Backend_1(8), jvm_TxInjector_1(8)
|
OS Image Description |
os_1
|
Tuning |
- ulimit -n 655350
- tuned-adm profile throughput-performance
- echo 16000000 > /proc/sys/kernel/sched_latency_ns
- echo 80000000 > /proc/sys/kernel/sched_min_granularity_ns
- echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
- echo 50000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
- echo 990000 > /proc/sys/kernel/sched_rt_runtime_us
- echo 16 > /proc/sys/kernel/sched_nr_migrate
- echo 10 > /proc/sys/vm/swappiness
- echo 10000 > /proc/sys/vm/dirty_expire_centisecs
- echo 1500 > /proc/sys/vm/dirty_writeback_centisecs
- echo 40 > /proc/sys/vm/dirty_ratio
- echo 10 > /proc/sys/vm/dirty_background_ratio
|
Notes |
None
|
Parts of Benchmark |
Controller
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m -XX:+UseParallelOldGC -XX:ParallelGCThreads=2 -XX:CICompilerCount=2
|
Tuning |
numactl --interleave=all
|
Notes |
None
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
-showversion -server -Xmx29g -Xms29g -Xmn28g -XX:+UseParallelOldGC -XX:ParallelGCThreads=32 -XX:+UseLargePages -XX:LargePageSizeInBytes=2m -XX:+UseBiasedLocking -XX:+AggressiveOpts -XX:+AlwaysPreTouch -XX:-UseAdaptiveSizePolicy -XX:SurvivorRatio=28
|
Tuning |
numactl used to affinitize each Backend JVM to a single memory node, eg:- numactl --cpunodebind=$gnum --localalloc
|
Notes |
None
|
Parts of Benchmark |
TxInjector
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m -XX:+UseParallelOldGC -XX:ParallelGCThreads=3 -XX:CICompilerCount=2
|
Tuning |
numactl used to affinitize each Backend JVM to a single memory node, eg:- numactl --cpunodebind=$gnum --localalloc
|
Notes |
None
|