Super Micro Computer, Inc Supermicro SuperBlade A+ Blade Server SBA-4119S-C2N 2998803 SPECjbb2015-Distributed max-jOPS
1416967 SPECjbb2015-Distributed critical-jOPS
Tested by: Super Micro Computer, Inc Test Sponsor: Super Micro Computer, Inc Test location: San Jose, California Test date: March 25, 2020
SPEC license #: 001176 Hardware Availability: Apr-2020 Software Availability: Jan-2020 Publication: Fri Apr 10 16:18:41 EDT 2020
Benchmark Results Summary
 
Overall Throughput RT curve
Overall SUT (System Under Test) Description
VendorSuper Micro Computer, Inc
Vendor URLhttps://www.supermicro.com
System SourceSingle Supplier
System DesignationServer Rack
Total Systems1
All SUT Systems IdenticalYES
Total Nodes20
All Nodes IdenticalYES
Nodes Per System20
Total Chips20
Total Cores1280
Total Threads2560
Total Memory Amount (GB)10240
Total OS Images20
SW EnvironmentNon-virtual
 
Hardware hw_1
NameSupermicro SuperBlade A+ Blade Server SBA-4119S-C2N
VendorSuper Micro Computer, Inc
Vendor URLhttps://www.supermicro.com
AvailableApr-2020
ModelBH12SSi
Form Factor8U
CPU NameAMD EPYC 7702P
CPU Characteristics64 Core, 2.0 GHz, 256 MB L3 cache (Turbo Boost Technology up to 3.35 GHz)
Number of Systems1
Nodes Per System20
Chips Per System20
Cores Per System1280
Cores Per Chip64
Threads Per System2560
Threads Per Core2
Version1.0 03/23/2020
CPU Frequency (MHz)2000
Primary Cache32KB(I)+32KB(D) per core
Secondary Cache512KB (I+D) per core
Tertiary Cache256MB (I+D) on chip per chip
Other CacheNone
Disk20 x 1TB M.2 PCIe Gen3
File Systemxfs
Memory Amount (GB)10240
# and size of DIMM(s)160 x 64GB
Memory Details64GB 2Rx4 PC4-3200AA
# and type of Network Interface Cards (NICs)Onboard 1x25GbE dual-port NIC
Power Supply Quantity and Rating (W)6 x 2200W
Other HardwareNone
Cabinet/Housing/EnclosureNone
Shared DescriptionNone
Shared CommentNone
Notes
  • NA: The test sponsor attests, as of date of publication, that CVE-2017-5754 (Meltdown) is mitigated in the system as tested and documented.
  • Yes: The test sponsor attests, as of date of publication, that CVE-2017-5753 (Spectre variant 1) is mitigated in the system as tested and documented.
  • Yes: The test sponsor attests, as of date of publication, that CVE-2017-5715 (Spectre variant 2) is mitigated in the system as tested and documented.
Other Hardware network_1
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Operating System os_1
NameRed Hat Enterprise Linux 7.7 (Maipo)
VendorRed Hat
Vendor URLhttp://www.redhat.com/
Version3.10.0-1062.el7.x86_64
AvailableAug-2019
Bitness64
NotesNone
Java Virtual Machine jvm_1
NameOracle Java SE 13.0.2
VendorOracle
Vendor URLhttp://oracle.com/
VersionJava HotSpot 64-bit Server VM, version 13.0.2
AvailableJan-2020
Bitness64
NotesNone
Other Software other_1
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Hardware hw_2
NameSupermicro A+ Server 2014TP-HTR
VendorSuper Micro Computer, Inc
Vendor URLhttps://www.supermicro.com
AvailableAug-2019
ModelH12SST-PS
Form Factor2U
CPU NameAMD EPYC 7702P
CPU Characteristics64 Core, 2.0 GHz, 256 MB L3 cache (Turbo Boost Technology up to 3.35 GHz)
Number of Systems1
Nodes Per System2
Chips Per System2
Cores Per System128
Cores Per Chip64
Threads Per System256
Threads Per Core2
Version1.0c 11/25/2019
CPU Frequency (MHz)2000
Primary Cache32KB(I)+32KB(D) per core
Secondary Cache512KB (I+D) per core
Tertiary Cache256MB (I+D) on chip per chip
Other CacheNone
Disk2 x SATA 240GB SSD
File Systemxfs
Memory Amount (GB)1024
# and size of DIMM(s)16 x 64GB
Memory Details2RX4 PC4-3200AA
# and type of Network Interface Cards (NICs)1x Supermicro AOC-MH25G-m2S2T Dual Port 25GbE SFP28 NIC
Power Supply Quantity and Rating (W)1 x 2000W
Other HardwareNone
Cabinet/Housing/EnclosureNone
Shared DescriptionNone
Shared CommentNone
Notes
  • NA: The test sponsor attests, as of date of publication, that CVE-2017-5754 (Meltdown) is mitigated in the system as tested and documented.
  • Yes: The test sponsor attests, as of date of publication, that CVE-2017-5753 (Spectre variant 1) is mitigated in the system as tested and documented.
  • Yes: The test sponsor attests, as of date of publication, that CVE-2017-5715 (Spectre variant 2) is mitigated in the system as tested and documented.
Other Hardware network_2
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Operating System os_2
NameRed Hat Enterprise Linux 7.7 (Maipo)
VendorRed Hat
Vendor URLhttp://www.redhat.com/
Version3.10.0-1062.el7.x86_64
AvailableAug-2019
Bitness64
NotesNone
Java Virtual Machine jvm_2
NameOracle Java SE 13.0.2
VendorOracle
Vendor URLhttp://oracle.com/
VersionJava HotSpot 64-bit Server VM, version 13.0.2
AvailableJan-2020
Bitness64
NotesNone
Other Software other_2
NameNone
VendorNone
Vendor URLNone
VersionNone
AvailableNone
BitnessNone
NotesNone
Hardware
OS Images os_Image_1(1)
Hardware Description hw_1
Number of Systems 1
SW Environment non-virtual
Tuning

BIOS Settings:

  • NUMA nodes per socket = NPS4
  • Determinism Control = Manual
  • Determinism Slider = Power
  • Memory Clock Speed = 1467MHz
  • L1 Stream HW Prefetcher = Disable
  • L2 Stream HW Prefetcher = Disable

Notes None
OS Image os_Image_1
JVM Instances jvm_Backend_1(80)
OS Image Description os_1
Tuning

  • cpupower -c all frequency-set -g performance
  • tuned-adm profile throughput-performance

  • echo 10000 > /proc/sys/kernel/sched_cfs_bandwidth_slice_us
  • echo 0 > /proc/sys/kernel/sched_child_runs_first
  • echo 16000000 > /proc/sys/kernel/sched_latency_ns
  • echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
  • echo 28000000 > /proc/sys/kernel/sched_min_granularity_ns
  • echo 9 > /proc/sys/kernel/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 > /proc/sys/kernel/sched_tunable_scaling
  • echo 50000000 > /proc/sys/kernel/sched_wakeup_granularity_ns

  • echo 3000 > /proc/sys/vm/dirty_expire_centisecs
  • echo 500 > /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 always > /sys/kernel/mm/transparent_hugepage/defrag
  • echo always > /sys/kernel/mm/transparent_hugepage/enabled

  • Add cgroup_disable=memory,cpu,cpuacct,blkio,hugetlb,pids,cpuset,perf_event,freezer,devices,net_cls,net_prio to GRUB_CMDLINE_LINUX_DEFAULT
  • ulimit -n 1024000
  • UserTasksMax=970000
  • DefaultTasksMax=970000

Notes None
JVM Instance jvm_Backend_1
Parts of Benchmark Backend
JVM Instance Description jvm_1
Command Line

-Xms32736M -Xmx32736M -Xmn31736M -XX:-UseDynamicNumberOfGCThreads -XX:CICompilerCount=32 -XX:AllocatePrefetchInstr=2 -XX:+UseParallelOldGC -XX:ParallelGCThreads=32 -XX:LargePageSizeInBytes=2m -XX:-UseAdaptiveSizePolicy -XX:+AlwaysPreTouch -XX:-UseBiasedLocking -XX:+UseLargePages -XX:TLABAllocationWeight=55 -XX:ThreadStackSize=512 -XX:SurvivorRatio=31 -XX:TargetSurvivorRatio=95 -XX:InlineSmallCode=10k -XX:MaxGCPauseMillis=300 -XX:LoopUnrollLimit=100 -XX:+UseTransparentHugePages

Tuning None
Notes

Used numactl to affinitize each Backend JVM to one NUMA node in every BH12SSi node in SuT

  • Group1:numactl --cpunodebind=0 --membind=0
  • Group2:numactl --cpunodebind=1 --membind=1
  • Group3:numactl --cpunodebind=2 --membind=2
  • Group4:numactl --cpunodebind=3 --membind=3

Hardware
OS Images os_Image_2(1)
Hardware Description hw_2
Number of Systems 1
SW Environment non-virtual
Tuning

BIOS Settings:

  • NUMA nodes per socket = NPS4
  • Determinism Control = Manual
  • Determinism Slider = Power
  • Memory Clock Speed = 1467MHz
  • L1 Stream HW Prefetcher = Disable
  • L2 Stream HW Prefetcher = Disable

Notes None
OS Image os_Image_2
JVM Instances jvm_Ctr_1(1), jvm_TxInjector_1(80)
OS Image Description os_2
Tuning

  • cpupower -c all frequency-set -g performance
  • tuned-adm profile throughput-performance

  • echo 10000 > /proc/sys/kernel/sched_cfs_bandwidth_slice_us
  • echo 0 > /proc/sys/kernel/sched_child_runs_first
  • echo 16000000 > /proc/sys/kernel/sched_latency_ns
  • echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
  • echo 28000000 > /proc/sys/kernel/sched_min_granularity_ns
  • echo 9 > /proc/sys/kernel/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 > /proc/sys/kernel/sched_tunable_scaling
  • echo 50000000 > /proc/sys/kernel/sched_wakeup_granularity_ns

  • echo 3000 > /proc/sys/vm/dirty_expire_centisecs
  • echo 500 > /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 always > /sys/kernel/mm/transparent_hugepage/defrag
  • echo always > /sys/kernel/mm/transparent_hugepage/enabled

  • Add cgroup_disable=memory,cpu,cpuacct,blkio,hugetlb,pids,cpuset,perf_event,freezer,devices,net_cls,net_prio to GRUB_CMDLINE_LINUX_DEFAULT
  • ulimit -n 1024000
  • UserTasksMax=970000
  • DefaultTasksMax=970000

Notes None
JVM Instance jvm_Ctr_1
Parts of Benchmark Controller
JVM Instance Description jvm_2
Command Line

-Xms6g -Xmx6g -Xmn5g -XX:+UseParallelOldGC -XX:CICompilerCount=12 -XX:ParallelGCThreads=12

Tuning None
Notes

Used numactl to interleave memory on all Nodes in Driver system

  • numactl --interleave=all

JVM Instance jvm_TxInjector_1
Parts of Benchmark TxInjector
JVM Instance Description jvm_1
Command Line

-Xms4g -Xmx4g -Xmn3g -XX:+UseParallelOldGC -XX:CICompilerCount=2 -XX:ParallelGCThreads=2

Tuning None
Notes

Used numactl to affinitize 20 Transaction Injector JVM to one NUMA node in Driver system

  • Group1:numactl --cpunodebind=0 --membind=0
  • Group2:numactl --cpunodebind=0 --membind=0
  • Group3:numactl --cpunodebind=0 --membind=0
  • ...
  • Group20:numactl --cpunodebind=0 --membind=0
  • Group21:numactl --cpunodebind=1 --membind=1
  • Group22:numactl --cpunodebind=1 --membind=1
  • ...
  • Group40:numactl --cpunodebind=1 --membind=1
  • ...
  • Group60:numactl --cpunodebind=2 --membind=2
  • ...
  • Group80:numactl --cpunodebind=3 --membind=3

max-jOPS = jOPS passed before the First Failure
Pass/Fail Pass Pass Fail Fail Fail
jOPS 2959345 2998803 3038261 3077719 3117177
critical-jOPS = Geomean ( jOPS @ 10000; 25000; 50000; 75000; 100000; SLAs )
Response time percentile is 99-th
SLA (us) 10000 25000 50000 75000 100000 Geomean
jOPS 1006177 1203467 1440215 1676962 1953168 1416967
  Percentile
  10-th 50-th 90-th 95-th 99-th 100-th
500us - / 39458 - / 39458 - / 39458 - / 39458 - / 39458 - / 39458
1000us 473495 / 512953 118374 / 157832 - / 39458 - / 39458 - / 39458 - / 39458
5000us 2959345 / 2998803 2249103 / 2288560 1223196 / 1262654 986448 / 1025906 552411 / 473495 - / 39458
10000us 2998803 / - 2722598 / 2762056 2012355 / 2051813 1815065 / 1854523 986448 / 1025906 - / 39458
25000us 2998803 / - 2840972 / 2880430 2525308 / 2564766 2288560 / 2328018 1183738 / 1223196 - / 39458
50000us 2998803 / - 2880430 / 2919887 2604224 / 2643682 2367476 / 2406934 1420486 / 1459944 39458 / 78916
75000us 2998803 / - 2919887 / 2959345 2643682 / 2683140 2446392 / 2485850 1657233 / 1696691 39458 / 78916
100000us 2998803 / - 2919887 / 2959345 2683140 / 2722598 2525308 / 2564766 1933439 / 1972897 39458 / 78916
200000us 2998803 / - 2998803 / - 2801514 / 2840972 2722598 / 2762056 2564766 / 2604224 789159 / 197290
500000us 2998803 / - 2998803 / - 2880430 / 2919887 2880430 / 2919887 2801514 / 2840972 2485850 / 2525308
1000000us 2998803 / - 2998803 / - 2959345 / 2998803 2919887 / 2959345 2880430 / 2919887 2762056 / 2801514
Probes jOPS / Total jOPS
Request Mix Accuracy
Note
(Actual % in the Mix - Expected % in the Mix) must be within:
'Main Tx' limit of +/-5.0% for the requests whose expected % in the mix is >= 10.0%
'Minor Tx' limit of +/-1.0% for the requests whose expected % in the mix is < 10.0%
There were no non-critical failures in Response Time curve building
Delay between status pings
IR/PR Accuracy
This section lists properties only set by user
Property Name Default Controller
specjbb.comm.connect.client.pool.size 256 210
specjbb.comm.connect.selector.runner.count 0 3
specjbb.comm.connect.timeouts.connect 60000 600000
specjbb.comm.connect.timeouts.read 60000 600000
specjbb.comm.connect.timeouts.write 60000 600000
specjbb.comm.connect.worker.pool.max 256 90
specjbb.controller.host localhost 172.27.33.243
specjbb.customerDriver.threads 64 {probe=64, saturate=64}
specjbb.forkjoin.workers 128 {Tier1=160, Tier2=14, Tier3=28}
specjbb.group.count 1 80
specjbb.heartbeat.period 10000 100000
specjbb.heartbeat.threshold 100000 1000000
specjbb.mapreducer.pool.size 128 28
specjbb.txi.pergroup.count 1 1
View table in csv format
 
Level: COMPLIANCE
Check Agent Result
Check properties on compliance All PASSED
 
Level: CORRECTNESS
Check Agent Result
Compare SM and HQ Inventory All PASSED
High-bound (max attempted) is 3945794 IR
High-bound (settled) is 3339028 IR