SPECpower_ssj2008

Fujitsu FUJITSU Server PRIMERGY CX2560 M5			SPECpower_ssj2008 = 10,619 overall ssj_ops/watt
Test Sponsor:	Fujitsu	SPEC License #:	19	Test Method:	Multi Node
Tested By:	Fujitsu	Test Location:	Kawasaki, Kanagawa, Japan	Test Date:	Apr 6, 2019
Hardware Availability:	Jun-2019	Software Availability:	Dec-2018	Publication:	May 22, 2019
System Source:	Single Supplier	System Designation:	Server	Power Provisioning:	Line-powered

Benchmark Results Summary

Performance			Power	Performance to Power Ratio
Target Load	Actual Load	ssj_ops	Average Active Power (W)	Performance to Power Ratio
100%	99.7%	17,566,126	1,412	12,441
90%	90.1%	15,861,908	1,214	13,061
80%	80.0%	14,094,424	1,078	13,076
70%	70.9%	12,493,145	969	12,894
60%	60.0%	10,566,162	867	12,188
50%	50.0%	8,798,423	787	11,183
40%	40.0%	7,042,265	718	9,814
30%	30.0%	5,279,188	652	8,092
20%	20.0%	3,523,498	587	6,005
10%	10.0%	1,758,976	520	3,380
Active Idle		0	329	0
∑ssj_ops / ∑power =				10,619

Aggregate SUT Data

# of Nodes	# of Chips	# of Cores	# of Threads	Total RAM (GB)	# of OS Images	# of JVM Instances
4	8	192	384	768	4	192

System Under Test

Shared Hardware

Shared Hardware
Enclosure:	PRIMERGY CX400 M4
Form Factor:	2U
Power Supply Quantity and Rating (W):	1 x 2400
Power Supply Details:	Fujitsu Technology Solutions S26113-F620-E30
Network Switch:	Fujitsu SR-X324T2
Network Switch Details:	24-Port Gigabit Switch
KVM Switch:	None
KVM Switch Details:	N/A
Other Hardware:	None
Comment:	Gigabit switch was not measured for power

Set: 'CX2560M5'

Set Identifier:	CX2560M5
Set Description:	System Under Test
# of Identical Nodes:	4
Comment:	Multi Node

Hardware per Node
Hardware Vendor:	Fujitsu
Model:	FUJITSU Server PRIMERGY CX2560 M5
Form Factor:	Blade
CPU Name:	Intel Xeon Platinum 6252
CPU Characteristics:	24-Core, 2.10GHz, 35.75MB L3 Cache
CPU Frequency (MHz):	2100
CPU(s) Enabled:	48 cores, 2 chips, 24 cores/chip
Hardware Threads:	96 (2 / core)
CPU(s) Orderable:	1,2 chips
Primary Cache:	32 KB I + 32 KB D on chip per core
Secondary Cache:	1 MB I+D on chip per core
Tertiary Cache:	36608 KB I+D on chip per chip
Other Cache:	None
Memory Amount (GB):	192
# and size of DIMM:	12 x 16384 MB
Memory Details:	12 x 16 GB 2Rx8 PC4-2933Y-R; slots 1A, 1B, 1C, 1D, 1E, 1F, 1G, 1H, 1J, 1K, 1L, 1M populated
Power Supply Quantity and Rating (W):	None
Power Supply Details:	N/A
Disk Drive:	1 x SSD M.2 SATA 6Gbps 128GB N H-P, S26361-F5658-L128
Disk Controller:	Integrated SATA Controller
# and type of Network Interface Cards (NICs) Installed:	1 x Intel(R) I210 Gigabit Network Connection (onboard)
NICs Enabled in Firmware / OS / Connected:	1/1/1
Network Speed (Mbit):	1000
Keyboard:	None
Mouse:	None
Monitor:	None
Optical Drives:	No
Other Hardware:	None

Software per Node
Power Management:	Enabled (see SUT Notes)
Operating System (OS):	SUSE Linux Enterprise Server 12 SP4
OS Version:	4.12.14-94.41-default
Filesystem:	xfs
JVM Vendor:	Oracle Corporation
JVM Version:	Oracle Java HotSpot(TM) 64-Bit Server VM (build 24.80-b11, mixed mode), version 1.7.0_80
JVM Command-line Options:	server -Xmn1700m -Xms1950m -Xmx1950m -XX:SurvivorRatio=1 -XX:TargetSurvivorRatio=99 -XX:AllocatePrefetchDistance=256 -XX:AllocatePrefetchLines=4 -XX:LoopUnrollLimit=45 -XX:InitialTenuringThreshold=12 -XX:MaxTenuringThreshold=15 -XX:ParallelGCThreads=8 -XX:InlineSmallCode=3900 -XX:MaxInlineSize=270 -XX:FreqInlineSize=2500 -XX:+AggressiveOpts -XX:+UseLargePages -XX:+UseParallelOldGC -XX:+UseHugeTLBFS -XX:+UseTransparentHugePages
JVM Affinity:	chrt -r 1 taskset -c [(0,48),(1,49),(2,50),(3,51),(7,55),(8,56),(9,57),(13,61),(14,62),(15,63),(19,67),(20,68),(4,52),(5,53),(6,54),(10,58),(11,59),(12,60),(16,64),(17,65),(18,66),(21,69),(22,70),(23,71),(24,72),(25,73),(26,74),(27,75),(31,79),(32,80),(33,81),(37,85),(38,86),(39,87),(43,91),(44,92),(28,76),(29,77),(30,78),(34,82),(35,83),(36,84),(40,88),(41,89),(42,90),(45,93),(46,94),(47,95)] numactl --membind=[0-3]
JVM Instances:	48
JVM Initial Heap (MB):	1950
JVM Maximum Heap (MB):	1950
JVM Address Bits:	64
Boot Firmware Version:	R1.6.0
Management Firmware Version:	2.41P
Workload Version:	SSJ 1.2.10
Director Location:	Controller
Other Software:	None

Boot Firmware Settings

Set "HWPM = Native Mode" in BIOS.
Set "ASPM Support = L1 Only" in BIOS.
Set "SATA Controller = Disable" in BIOS.
Set "USB Port Control = Disable all ports" in BIOS.
Set "Network Stack = Disabled" in BIOS.
Set "Hardware Prefetcher = Disabled" in BIOS.
Set "Adjacent Cache Line Prefetcher = Disabled" in BIOS.
Set "DCU Streamer Prefetcher = Disabled" in BIOS.
Set "Intel Virtualization Technology = Disabled" in BIOS.
Set "Power Technology = Custom" in BIOS.
Set "Turbo Mode = Disabled" in BIOS.
Set "Energy Performance = Energy Efficient" in BIOS.
Set "Override OS Energy Performance = Enabled" in BIOS.
Set "P-State Coordination = SW_ANY" in BIOS.
Set "Package C State Limit = C6" in BIOS.
Set "UPI Link Frequency Select = 9.6GT/s" in BIOS.
Set "Uncore Frequency Scaling = Disabled" in BIOS.
Set "Sub NUMA Clustering = Enabled" in BIOS.
Set "DDR Performance = Energy optimized" in BIOS.

Management Firmware Settings

None

System Under Test Notes

kernal parameter:pcie_aspm=force pcie_aspm.policy=powersave intel_pstate=disable rcu_nocbs=1-95 nohz=off isolcpus=1-95
Benchmark started via ssh
modprobe cpufreq_conservative
cpupower frequency-set --governor conservative
echo -n 98 > /sys/devices/system/cpu/cpufreq/conservative/up_threshold
echo -n 1 > /sys/devices/system/cpu/cpufreq/conservative/freq_step
echo -n 1000000 > /sys/devices/system/cpu/cpufreq/conservative/sampling_rate
echo -n 0 > /sys/devices/system/cpu/cpufreq/conservative/ignore_nice_load
sysctl -w kernel.sched_migration_cost_ns=6000
echo -n 97 > /sys/devices/system/cpu/cpufreq/conservative/down_threshold
echo -n 1 > /sys/devices/system/cpu/cpufreq/conservative/sampling_down_factor
sysctl -w kernel.sched_min_granularity_ns=10000000
echo always > /sys/kernel/mm/transparent_hugepage/enabled
powertop --auto-tune
echo 0 > /proc/sys/kernel/nmi_watchdog
sysctl -w vm.swappiness=50
sysctl -w vm.laptop_mode=5
<Yes>: 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.

Controller System

Hardware
Hardware Vendor:	Fujitsu
Model:	PRIMERGY RX2530 M2
CPU Description:	Intel Xeon E5-2643 v4
Memory amount (GB):	64

Software
Operating System (OS):	Microsoft Windows Server 2012 R2 Standard
JVM Vendor:	Oracle Corporation
JVM Version:	Oracle Java HotSpot(TM) 64-Bit Server VM (build 24.80-b11, mixed mode), version 1.7.0_80
CCS Version:	1.2.6

Measurement Devices

Power Analyzer pwr1
Hardware Vendor:	Hioki
Model:	Hioki3336
Serial Number:	170174584
Connectivity:	RS-232
Input Connection:	Default
Metrology Institute:	NICT (National Institute of Information and Communications Technology)
Accredited by:	HIOKI E.E. CORPORATION
Calibration Label:	H06400087-1901
Date of Calibration:	30-Jan-2019
PTDaemon Host System:	same as CCS
PTDaemon Host OS:	same as CCS
PTDaemon Version:	1.8.1-260d064e-20161220
Setup Description:	Connected to PSU 1

Temperature Sensor temp1
Hardware Vendor:	Digi International Inc
Model:	Watchport/H
Driver Version:	Watchport Virtual Port 5.70.105.0
Connectivity:	USB
PTDaemon Host System:	same as CCS
PTDaemon Host OS:	same as CCS
Setup Description:	5 mm in front of SUT main air intake

Notes

SPECpower_ssj.props input.load_level.number_warehouses set to 96 due to a known inconsistency in processor reporting with this Java version.

Aggregate Electrical and Environmental Data

Target Load	Average Active Power (W)	Minimum Ambient Temperature (°C)
100%	1,412	21.1
90%	1,214	21.1
80%	1,078	21.2
70%	969	21.2
60%	867	21.2
50%	787	21.1
40%	718	21.1
30%	652	21.1
20%	587	20.9
10%	520	20.9
Active Idle	329	20.9

Line Standard	Minimum Temperature (°C)	Elevation (m)
200V / 50 Hz / 1 phase / 2 wires	20.9	11

See the Power/Temperature Details Report for additional details.

Aggregate Performance Data

Target Load	Actual Load	ssj_ops
Target Load	Actual Load	Target	Actual
Calibration 1			17,470,634
Calibration 2			17,613,693
Calibration 3			17,608,389
ssj_ops@calibrated=17,611,041
100%	99.7%	17,611,041	17,566,126
90%	90.1%	15,849,937	15,861,908
80%	80.0%	14,088,833	14,094,424
70%	70.9%	12,327,729	12,493,145
60%	60.0%	10,566,625	10,566,162
50%	50.0%	8,805,521	8,798,423
40%	40.0%	7,044,417	7,042,265
30%	30.0%	5,283,312	5,279,188
20%	20.0%	3,522,208	3,523,498
10%	10.0%	1,761,104	1,758,976
Active Idle		0	0

See the Aggregate Performance Report for additional details.

http://www.spec.org - info@spec.org

SPECpower_ssj2008 Reporter Version: [SSJ 1.2.10, May 9, 2012]