IBM Corporation IBM Cloud AMD EPYC 7F72 Bare Metal Server 165668 SPECjbb2015-MultiJVM max-jOPS
109720 SPECjbb2015-MultiJVM critical-jOPS
Tested by: IBM Corporation Test Sponsor: IBM Corporation Test location: Austin, TX Test date: Apr 21, 2020
SPEC license #: 11 Hardware Availability: Jun-2020 Software Availability: Jan-2020 Publication: Tue May 19 11:13:37 EDT 2020
Benchmark Results Summary
 
Overall Throughput RT curve
Overall SUT (System Under Test) Description
VendorIBM Corporation
Vendor URLhttps://cloud.ibm.com
System SourceSingle Supplier
System DesignationServer Rack
Total Systems1
All SUT Systems IdenticalYES
Total Nodes1
All Nodes IdenticalYES
Nodes Per System1
Total Chips2
Total Cores48
Total Threads96
Total Memory Amount (GB)512
Total OS Images1
SW EnvironmentNon-virtual
 
Hardware hw_1
NameIBM Cloud AMD EPYC 7F72 Bare Metal Server
VendorIBM Corporation
Vendor URLhttps://cloud.ibm.com
AvailableJun-2020
ModelNone
Form Factor2U
CPU NameAMD EPYC 7F72
CPU Characteristics24 Core, 3.20 GHz, 192 MB L3 cache (Turbo Boost Technology up to 3.70 GHz)
Number of Systems1
Nodes Per System1
Chips Per System2
Cores Per System48
Cores Per Chip24
Threads Per System96
Threads Per Core2
Version1.0 03/25/2020
CPU Frequency (MHz)3200
Primary Cache32KB(I)+32KB(D) per core
Secondary Cache512KB (I+D) per core
Tertiary Cache192MB (I+D) on chip per chip
Other CacheNone
Disk1x960 GB SSD
File Systemext4
Memory Amount (GB)512
# and size of DIMM(s)16 x 32GB
Memory Details32 GB 2Rx4 PC4-2933-R
# and type of Network Interface Cards (NICs)4 x 10 GbE
Power Supply Quantity and Rating (W)2 x 1600W
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
    NameUbuntu 18.04.4 LTS
    VendorUbuntu
    Vendor URLhttp://ubuntu.com/
    Version4.15.0-91-generic
    AvailableFeb-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
    OS Images os_Image_1(1)
    Hardware Description hw_1
    Number of Systems 1
    SW Environment non-virtual
    Tuning

    BIOS Settings:

    • Set Nodes per Socket to 2 [NPS2]
    • SMT Control = Auto [Auto means SMT Enabled]
    • Determinism Control = Manual
    • Determinism Slider = Power
    • L1 Stream HW Prefetcher = Disable
    • L2 Stream HW Prefetcher = Disable

    Notes None
    OS Image os_Image_1
    JVM Instances jvm_Ctr_1(1), jvm_Backend_1(4), jvm_TxInjector_1(4)
    OS Image Description os_1
    Tuning

    • cpupower -c all frequency-set -g performance
    • tuned-adm profile latency-performance
    • echo 10000000 > /proc/sys/kernel/sched_min_granularity_ns
    • echo 15000000 > /proc/sys/kernel/sched_wakeup_granularity_ns
    • echo 1000 > /proc/sys/kernel/sched_migration_cost_ns
    • echo 990000 > /proc/sys/kernel/sched_rt_runtime_us
    • echo 24000000 > /proc/sys/kernel/sched_latency_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
    • echo 0 > /proc/sys/kernel/numa_balancing
    • echo always > /sys/kernel/mm/transparent_hugepage/enabled
    • echo always > /sys/kernel/mm/transparent_hugepage/defrag
    • ulimit -n 1024000
    • UserTasksMax=970000
    • DefaultTasksMax=970000

    Notes None
    JVM Instance jvm_Ctr_1
    Parts of Benchmark Controller
    JVM Instance Description jvm_1
    Command Line

    -Xms2g -Xmx2g -Xmn1536m -XX:+UseParallelOldGC -XX:ParallelGCThreads=2

    Tuning None
    Notes

    Used numactl to interleave memory on all Nodes in Driver system

    • numactl --interleave=all

    JVM Instance jvm_Backend_1
    Parts of Benchmark Backend
    JVM Instance Description jvm_1
    Command Line

    -Xms118g -Xmx118g -Xmn116g -server -XX:MetaspaceSize=30m -XX:AllocatePrefetchInstr=2 -XX:LargePageSizeInBytes=2m -XX:-UsePerfData -XX:-UseAdaptiveSizePolicy -XX:+AlwaysPreTouch -XX:-UseBiasedLocking -XX:+UseLargePages -XX:+UseParallelOldGC -XX:SurvivorRatio=42 -XX:TargetSurvivorRatio=95 -XX:ParallelGCThreads=24 -XX:MaxTenuringThreshold=15 -XX:InitialCodeCacheSize=25m -XX:+UseCompressedOops -XX:ObjectAlignmentInBytes=32 -XX:+UseTransparentHugePages

    Tuning

    Used numactl to affinitize each Backend JVM to one NUMA node

    • numactl --cpunodebind=0 --membind=0
    • numactl --cpunodebind=1 --membind=1
    • numactl --cpunodebind=2 --membind=2
    • numactl --cpunodebind=3 --membind=3

    Notes "Notes here"
    JVM Instance jvm_TxInjector_1
    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 one NUMA node

    • numactl --cpunodebind=0 --membind=0
    • numactl --cpunodebind=1 --membind=1
    • numactl --cpunodebind=2 --membind=2
    • numactl --cpunodebind=3 --membind=3
    max-jOPS = jOPS passed before the First Failure
    Pass/Fail Pass Pass Fail Fail Fail
    jOPS 163301 165668 168034 170401 172768
    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 67450 113009 117150 128984 138056 109720
      Percentile
      10-th 50-th 90-th 95-th 99-th 100-th
    500us 2367 / 4733 - / 2367 - / 2367 - / 2367 - / 2367 - / 2367
    1000us 18933 / 21300 4733 / 7100 2367 / 4733 2367 / 4733 - / 2367 - / 2367
    5000us 163301 / 165668 149101 / 151468 71000 / 73367 42600 / 40234 23667 / 26033 4733 / 7100
    10000us 165668 / - 153834 / 156201 130167 / 132534 108867 / 111234 66267 / 68634 26033 / 11833
    25000us 165668 / - 160934 / 163301 146734 / 149101 139634 / 142001 118334 / 111234 26033 / 11833
    50000us 165668 / - 163301 / 165668 153834 / 156201 146734 / 149101 118334 / 115967 26033 / 11833
    75000us 165668 / - 165668 / - 156201 / 158568 149101 / 151468 130167 / 127801 26033 / 11833
    100000us 165668 / - 165668 / - 156201 / 158568 153834 / 151468 139634 / 134901 61534 / 11833
    200000us 165668 / - 165668 / - 160934 / 163301 158568 / 160934 149101 / 151468 115967 / 97034
    500000us 165668 / - 165668 / - 165668 / - 165668 / - 165668 / - 153834 / 151468
    1000000us 165668 / - 165668 / - 165668 / - 165668 / - 165668 / - 165668 / 163301
    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 Group1.Backend.beJVM Group1.TxInjector.txiJVM1 Group2.Backend.beJVM Group2.TxInjector.txiJVM1 Group3.Backend.beJVM Group3.TxInjector.txiJVM1 Group4.Backend.beJVM Group4.TxInjector.txiJVM1
    specjbb.comm.connect.client.pool.size 256 256 64 64 64 64 64 64 64 64
    specjbb.comm.connect.selector.runner.count 0 1
    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 256 256 64 64 64 64 64 64 64 64
    specjbb.controller.handshake.period 5000 20000
    specjbb.controller.handshake.timeout 600000 800000
    specjbb.forkjoin.workers 96 {Tier1=72, Tier2=4, Tier3=8}
    specjbb.group.count 1 4
    specjbb.heartbeat.period 10000 2000
    specjbb.heartbeat.threshold 100000 600000
    specjbb.mapreducer.pool.size 96 4
    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 236668 IR
    High-bound (settled) is 197290 IR