|
|
|
| JVM Instances |
jvm_Ctr_1(1), jvm_Backend_1(16), jvm_TxInjector_1(16)
|
| OS Image Description |
os_1
|
| Tuning |
- tuned-adm profile=throughput-performance
- pmqos-static.py cpu_dma_latency=133
- ulimit -n 65536
- 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 |
None
|
|
| 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=48 -XX:MaxTenuringThreshold=15 -Xnoclassgc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps
|
| Tuning |
None
|
| Notes |
Used numactl to affinitize each Backend JVM to a NUMA node. - numactl --physcpubind=0-23,384-407 --membind=0
- numactl --physcpubind=24-47,408-431 --membind=1
- numactl --physcpubind=48-71,432-455 --membind=2
- numactl --physcpubind=72-95,456-479 --membind=3
- numactl --physcpubind=96-119,480-503 --membind=4
- numactl --physcpubind=120-143,504-527 --membind=5
- numactl --physcpubind=144-167,528-551 --membind=6
- numactl --physcpubind=168-191,552-575 --membind=7
- numactl --physcpubind=192-215,576-599 --membind=8
- numactl --physcpubind=216-239,600-623 --membind=9
- numactl --physcpubind=240-263,624-647 --membind=10
- numactl --physcpubind=264-287,648-671 --membind=11
- numactl --physcpubind=288-311,672-695 --membind=12
- numactl --physcpubind=312-335,696-719 --membind=13
- numactl --physcpubind=336-359,720-743 --membind=14
- numactl --physcpubind=360-383,744-767 --membind=15
|
|
| 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=24 --membind=1
- numactl --physcpubind=48 --membind=2
- numactl --physcpubind=72 --membind=3
- numactl --physcpubind=96 --membind=4
- numactl --physcpubind=120 --membind=5
- numactl --physcpubind=144 --membind=6
- numactl --physcpubind=168 --membind=7
- numactl --physcpubind=192 --membind=8
- numactl --physcpubind=216 --membind=9
- numactl --physcpubind=240 --membind=10
- numactl --physcpubind=264 --membind=11
- numactl --physcpubind=288 --membind=12
- numactl --physcpubind=312 --membind=13
- numactl --physcpubind=336 --membind=14
- numactl --physcpubind=360 --membind=15
|
|
