JVM Instances |
jvm_Ctr_1(1), jvm_Backend_1(8), jvm_TxInjector_1(8)
|
OS Image Description |
os_1
|
Tuning |
- tuned-adm profile=throughput-performance
- echo 990000 > /proc/sys/kernel/sched_rt_runtime_us
- echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
- echo 16000000 > /proc/sys/kernel/sched_latency_ns
- echo 6000000 > /proc/sys/kernel/sched_min_granularity_ns
- echo 15000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
- 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
- echo 10 > /proc/sys/vm/swappiness
|
Notes |
None
|
Parts of Benchmark |
Controller
|
JVM Instance Description |
jvm_1
|
Command Line |
-Xms2g -Xmx2g -Xmn1536m -XX:+UseParallelOldGC -XX:ParallelGCThreads=1
|
Tuning |
None
|
Notes |
Used numactl to interleave memory on all nodes
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
-showversion -server -XX:AllocatePrefetchInstr=2 -XX:LargePageSizeInBytes=2m -XX:-UsePerfData -XX:+AggressiveOpts -XX:-UseAdaptiveSizePolicy -XX:+AlwaysPreTouch -XX:-UseBiasedLocking -XX:+UseLargePages -XX:+UseParallelOldGC -Xms29g -Xmx29g -Xmn27g -XX:SurvivorRatio=28 -XX:TargetSurvivorRatio=95 -XX:ParallelGCThreads=24 -XX:MaxTenuringThreshold=15 -Xnoclassgc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps
|
Tuning |
None
|
Notes |
Used numactl to affinitize each Backend JVM to physical cores in a NUMA node. - numactl --physcpubind=0-11,96-107 --membind=0
- numactl --physcpubind=12-23,108-119 --membind=1
- numactl --physcpubind=24-35,120-131 --membind=2
- numactl --physcpubind=36-47,132-143 --membind=3
- numactl --physcpubind=48-59,144-155 --membind=4
- numactl --physcpubind=60-71,156-167 --membind=5
- numactl --physcpubind=72-83,168-179 --membind=6
- numactl --physcpubind=84-95,180-191 --membind=7
|
Parts of Benchmark |
TxInjector
|
JVM Instance Description |
jvm_1
|
Command Line |
-Xms2g -Xmx2g -Xmn1536m -XX:+UseParallelOldGC -XX:ParallelGCThreads=2
|
Tuning |
None
|
Notes |
Used numactl to affinitize each Transaction Injector JVM to a physical core in a NUMA node. - numactl --physcpubind=0 --membind=0
- numactl --physcpubind=12 --membind=1
- numactl --physcpubind=24 --membind=2
- numactl --physcpubind=36 --membind=3
- numactl --physcpubind=48 --membind=4
- numactl --physcpubind=60 --membind=5
- numactl --physcpubind=72 --membind=6
- numactl --physcpubind=84 --membind=7
|