JVM Instances |
jvm_Ctr_1(1), jvm_Backend_1(4), jvm_TxInjector_1(4)
|
OS Image Description |
os_1
|
Tuning |
- tuned-adm profile throughput-performance
- cpupower frequency-set -g performance
- systemctl stop systemd-update-utmp-runlevel.service
- ulimit -n 655360
- echo 16000000 > /proc/sys/kernel/sched_latency_ns
- echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
- echo 155000 > /proc/sys/kernel/sched_min_granularity_ns
- echo 16 > /proc/sys/kernel/sched_nr_migrate
- echo 3000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
- echo always > /sys/kernel/mm/transparent_hugepage/defrag
- echo always > /sys/kernel/mm/transparent_hugepage/enabled
- ulimit -v 800000000
- ulimit -m 800000000
- ulimit -l 800000000
|
Notes |
None
|
Parts of Benchmark |
Controller
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m
|
Tuning |
Used numactl to interleave memory on all CPUs
|
Notes |
None
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms225g -Xmx225g -Xmn220g -XX:SurvivorRatio=34 -XX:MaxTenuringThreshold=15 -XX:+UseLargePages -XX:LargePageSizeInBytes=2m -XX:+UseParallelGC -Xnoclassgc -XX:+AlwaysPreTouch -XX:-UseAdaptiveSizePolicy -XX:-UsePerfData -XX:TargetSurvivorRatio=90 -XX:ParallelGCThreads=60 -XX:-UseBiasedLocking -XX:UseAVX=0
|
Tuning |
Used numactl to affinitize each Backend JVM to one NUMA node- numactl --cpunodebind=0 --localalloc
- numactl --cpunodebind=1 --localalloc
- numactl --cpunodebind=2 --localalloc
- numactl --cpunodebind=3 --localalloc
|
Notes |
None
|
Parts of Benchmark |
TxInjector
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m
|
Tuning |
Used numactl to affinitize each TxInjector JVM to one NUMA node- numactl --cpunodebind=0 --localalloc
- numactl --cpunodebind=1 --localalloc
- numactl --cpunodebind=2 --localalloc
- numactl --cpunodebind=3 --localalloc
|
Notes |
None
|