JVM Instances |
jvm_Ctr_1(1), jvm_Backend_1(4), jvm_TxInjector_1(4)
|
OS Image Description |
os_1
|
Tuning |
- ulimit -n 131072
- sysctl net.core.wmem_max=12582912
- sysctl net.core.rmem_max=12582912
- sysctl net.ipv4.tcp_rmem='10240 87380 12582912'
- sysctl net.ipv4.tcp_wmem='10240 87380 12582912'
- sysctl net.core.netdev_max_backlog=655560
- sysctl net.core.somaxconn=32768
- sysctl net.ipv4.tcp_no_metrics_save=1
- systemctl stop systemd-update-utmp-runlevel.service
- echo 10000 > /proc/sys/kernel/sched_cfs_bandwidth_slice_us
- echo 0 > /proc/sys/kernel/sched_child_runs_first
- echo 16000000 > /sys/kernel/debug/sched/latency_ns
- echo 1000 > /sys/kernel/debug/sched/migration_cost_ns
- echo 28000000 > /sys/kernel/debug/sched/min_granularity_ns
- echo 9 > /sys/kernel/debug/sched/nr_migrate
- echo 100 > /proc/sys/kernel/sched_rr_timeslice_ms
- echo 1000000 > /proc/sys/kernel/sched_rt_period_us
- echo 990000 > /proc/sys/kernel/sched_rt_runtime_us
- echo 0 > /proc/sys/kernel/sched_schedstats
- echo 1 > /sys/kernel/debug/sched/tunable_scaling
- echo 50000000 > /sys/kernel/debug/sched/wakeup_granularity_ns
- echo 0 > /proc/sys/kernel/numa_balancing
- 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
- tuned-adm profile throughput-performance
- echo 1500 > /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages
|
Notes |
None
|
Parts of Benchmark |
Controller
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m -XX:+UseLargePages -XX:LargePageSizeInBytes=1G -XX:+UseParallelGC -XX:ParallelGCThreads=2
|
Tuning |
None
|
Notes |
Used numactl to interleave memory on all CPUs
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
-showversion -XX:+UseParallelGC -XX:+UseLargePages -XX:+AlwaysPreTouch -XX:-UseAdaptiveSizePolicy -XX:MaxTenuringThreshold=15 -XX:InlineSmallCode=10k -XX:-UseCountedLoopSafepoints -XX:LoopUnrollLimit=20 -server -XX:TargetSurvivorRatio=99 -XX:SurvivorRatio=28 -XX:LargePageSizeInBytes=1g -XX:UseAVX=0 -Xms31744m -Xmx31744m -Xmn29696m -XX:ParallelGCThreads=30
|
Tuning |
numactl used to affinitize each Backend JVM to a single memory node, eg:- numactl --cpunodebind=$node --localalloc
|
Notes |
None
|
Parts of Benchmark |
TxInjector
|
JVM Instance Description |
jvm_1
|
Command Line |
-server -Xms2g -Xmx2g -Xmn1536m -XX:+UseLargePages -XX:LargePageSizeInBytes=1G -XX:+UseParallelGC -XX:ParallelGCThreads=2
|
Tuning |
numactl used to affinitize each TxInjector JVM to a single memory node, eg:- numactl --cpunodebind=$node --localalloc
|
Notes |
None
|