OS Images |
os_Image_1(1)
|
Hardware Description |
hw_1
|
Number of Systems |
1
|
SW Environment |
Non-virtual
|
Tuning |
- Dram Refresh Delay - Performance
- DIMM Self Healing (Post Package Repair) on Uncorrectable Memory Error - Disabled
- Correctable Error Logging - Disabled
- Virtualization Technology - Disabled
- L2 Stream HW Prefetcher - Disabled
- NUMA Nodes Per Socket - 4
- L3 cache as NUMA Domain- Enabled
- TDP Control- Manual
- Customized cTDP - 500
- Customized PPT - 500
- System Profile - Custom
- CPU Power Management - Maximum Performance
- C-States - Disabled
- Determinism Control- Manual
- Determinism Slider - Power Determinism
|
Notes |
None
|
|
JVM Instances |
jvm_Backend_1(24)
|
OS Image Description |
os_1
|
Tuning |
- ulimit -n 1024000
- echo 960000 > /proc/sys/kernel/sched_rt_runtime_us
- echo 999000000 > /proc/sys/kernel/sched_latency_ns
- echo 40000 > /proc/sys/kernel/sched_migration_cost_ns
- echo 999000000 > /proc/sys/kernel/sched_min_granularity_ns
- echo 999000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
- echo 9000 > /proc/sys/kernel/sched_nr_migrate
- 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
- echo 0 > /proc/sys/kernel/numa_balancing
- echo 0 > /proc/sys/vm/numa_stat
- echo always > /sys/kernel/mm/transparent_hugepage/enabled
- echo always > /sys/kernel/mm/transparent_hugepage/defrag
|
Notes |
None
|
Parts of Benchmark |
Backend
|
JVM Instance Description |
jvm_1
|
Command Line |
--Xms31g -Xmx31g -Xmn28g -XX:AllocatePrefetchInstr=2 -XX:+UseParallelGC -XX:ParallelGCThreads=32 -XX:LargePageSizeInBytes=2m -XX:-UseAdaptiveSizePolicy -XX:+AlwaysPreTouch -XX:+UseLargePages -XX:SurvivorRatio=35 -XX:TargetSurvivorRatio=95 -XX:MaxTenuringThreshold=15 -XX:InlineSmallCode=11k -XX:MaxGCPauseMillis=100 -XX:LoopUnrollLimit=200 -XX:+UseTransparentHugePages -XX:TLABAllocationWeight=2 -XX:CompileThresholdScaling=120 -XX:CICompilerCount=4 -XX:AutoBoxCacheMax=32 -XX:OnStackReplacePercentage=100 -XX:TLABSize=64k -XX:MinTLABSize=64k -XX:-ResizeTLAB -XX:TLABWasteTargetPercent=1 -XX:TLABWasteIncrement=1 -XX:YoungPLABSize=64k -XX:OldPLABSize=64k
|
Tuning |
Used numactl to affinitize each Backend JVM to 16 Core / 32 Threads- Group1: --physcpubind=0-15,384-399 --localalloc
- Group2: --physcpubind=16-31,400-415 --localalloc
- Group3: --physcpubind=32-47,416-431 --localalloc
- Group4: --physcpubind=48-63,432-447 --localalloc
- Group5: --physcpubind=64-79,448-463 --localalloc
- Group6: --physcpubind=80-95,464-479 --localalloc
- Group7: --physcpubind=96-111,480-495 --localalloc
- Group8: --physcpubind=112-127,496-511 --localalloc
- Group9: --physcpubind=128-143,512-527 --localalloc
- Group10: --physcpubind=144-159,528-543 --localalloc
- Group11: --physcpubind=160-175,544-559 --localalloc
- Group12: --physcpubind=176-191,560-575 --localalloc
- Group13: --physcpubind=192-207,576-591 --localalloc
- Group14: --physcpubind=208-223,592-607 --localalloc
- Group15: --physcpubind=224-239,608-623 --localalloc
- Group16: --physcpubind=240-255,624-639 --localalloc
- Group17: --physcpubind=256-271,640-655 --localalloc
- Group18: --physcpubind=272-287,656-671 --localalloc
- Group19: --physcpubind=288-303,672-687 --localalloc
- Group20: --physcpubind=304-319,688-703 --localalloc
- Group21: --physcpubind=320-335,704-719 --localalloc
- Group22: --physcpubind=336-351,720-735 --localalloc
- Group23: --physcpubind=352-367,736-751 --localalloc
- Group24: --physcpubind=368-383,752-767 --localalloc
|
Notes |
None
|
|