SPECpower_ssj2008

IBM Corporation IBM System x iDataPlex dx360 M4			SPECpower_ssj2008 = NC overall ssj_ops/watt
Test Sponsor:	IBM Corporation	SPEC License #:	11	Test Method:	Multi Node
Tested By:	IBM Corporation	Test Location:	Taipei, Taiwan	Test Date:	Sep 9, 2013
Hardware Availability:	Nov-2013	Software Availability:	Dec-2013	Publication:	Oct 2, 2013
System Source:	Single Supplier	System Designation:	Server	Power Provisioning:	Line-powered

SPEC has determined that this result is not in compliance with the SPECpower_ssj2008 run and reporting rules. SPEC has determined that this result is not in compliance with the SPECpower_ssj2008 Run & Reporting Rules 2.10 General Availability. Specifically the submitter generated the results whilst using a combination of Microsoft Windows Server 2012 or 2012 R2 together with the IBM J9 7.0 SR1 virtual machine. IBM has subsequently stated that this combination is unsupported.

Aggregate SUT Data

# of Nodes	# of Chips	# of Cores	# of Threads	Total RAM (GB)	# of OS Images	# of JVM Instances
2	4	40	80	48.0	2	20

System Under Test

Shared Hardware

Shared Hardware
Enclosure:	iDataPlex dx360 M4 2U Chassis
Form Factor:	2U
Power Supply Quantity and Rating (W):	1 x 750
Power Supply Details:	IBM 750W PSU (IBM P/N: 94Y5974)
Network Switch:	DLink DGS-1024D
Network Switch Details:	24-PORT GIGABIT DESKTOP SWITCH
KVM Switch:	N/A
KVM Switch Details:	N/A
Other Hardware:	None
Comment:	Network switch is a standalone device and is not included in the measured power.

Set: 'sut'

Set Identifier:	sut
Set Description:	System Under Test
# of Identical Nodes:	2
Comment:	None

Hardware per Node
Hardware Vendor:	IBM Corporation
Model:	IBM System x iDataPlex dx360 M4
Form Factor:	1U
CPU Name:	Intel Xeon E5-2660 v2
CPU Characteristics:	10-Core, 2.2GHz, 25MB L3 Cache
CPU Frequency (MHz):	2200
CPU(s) Enabled:	20 cores, 2 chips, 10 cores/chip
Hardware Threads:	40 (2 / core)
CPU(s) Orderable:	1,2 chips
Primary Cache:	32 KB I + 32 KB D on chip per core
Secondary Cache:	256 KB I+D on chip per core
Tertiary Cache:	25 MB I+D on chip per chip
Other Cache:	None
Memory Amount (GB):	24
# and size of DIMM:	6 x 4096 MB
Memory Details:	4GB 2Rx8 PC3L-12800E; slots 1, 3, 8, 9, 11, and 16 populated
Power Supply Quantity and Rating (W):	None
Power Supply Details:	N/A
Disk Drive:	1 x 128 GB SATA SSD (IBM P/N: 90Y8668)
Disk Controller:	Integrated SATA controller
# and type of Network Interface Cards (NICs) Installed:	2 x Integrated Intel I350 Gigabit Ethernet Controller
NICs Enabled in Firmware / OS / Connected:	2/2/1
Network Speed (Mbit):	1000
Keyboard:	None
Mouse:	None
Monitor:	None
Optical Drives:	No
Other Hardware:	None

Software per Node
Power Management:	IBM Power Plan for Windows Server 2012
Operating System (OS):	Microsoft Windows Server 2012 DataCenter Edition
OS Version:	Version 6.2.9200 Build 9200
Filesystem:	NTFS
JVM Vendor:	IBM Corporation
JVM Version:	IBM J9 VM (build 2.6, JRE 1.7.0 Windows Server 2008 R2 amd64-64 20120322_106209 (JIT enabled, AOT enabled)
JVM Command-line Options:	-Xaggressive -Xcompressedrefs -Xmn1100m -Xms1500m -Xmx1500m -XlockReservation -Xnoloa -XtlhPrefetch -Xlp -Xconcurrentlevel0
JVM Affinity:	start /affinity [F,F0,F00,F000,F0000,F00000,F000000,F0000000,F00000000,F000000000]
JVM Instances:	10
JVM Initial Heap (MB):	1500
JVM Maximum Heap (MB):	1500
JVM Address Bits:	64
Boot Firmware Version:	TDE133G
Management Firmware Version:	1AOO43C
Workload Version:	SSJ 1.2.10
Director Location:	Controller
Other Software:	IBM SDK Java Technology Edition Version 7.0 for Windows x64, IBM Power Plan for Windows Server 2012

Boot Firmware Settings

BIOS Settings
- Operating Mode: Custom
- Memory Power Management: Automatic
- QPI Link Frequency: Minimal Power
- Power/Performance Bias: OS Controlled
- Hardware Prefetcher: Disable
- Adjacent Cache Prefetch: Disable
- DCU Streamer Prefetcher: Disable
- Patrol Scrub: Disable
- Disabled Mezz_Card, Slot 1 and Slot 2
- Commands on USB interface: Disabled

Management Firmware Settings

None

System Under Test Notes

Each JVM instance was affinitized four threads on a single socket.
Using the local security settings console, "lock pages in memory" was enabled for the user running the benchmark.
Run was started on controller using Remote Desktop.

Controller System

Hardware
Hardware Vendor:	IBM Corporation
Model:	IBM System x iDataPlex dx320 Server
CPU Description:	Intel Xeon CPU L5420
Memory amount (GB):	24

Software
Operating System (OS):	Microsoft Windows Server 2008 R2 Enterprise Edition
JVM Vendor:	IBM Corporation
JVM Version:	IBM J9 VM (build pwa6460sr5-20090529_04(SR5))
CCS Version:	1.2.6

Measurement Devices

Power Analyzer pwr1
Hardware Vendor:	Yokogawa Meters and Instruments Corporation
Model:	WT210
Serial Number:	91JB14535
Connectivity:	RS-232
Input Connection:	Default
Metrology Institute:	NML/ROC
Accredited by:	ETC, Taiwan
Calibration Label:	13-04-BAC-294-03
Date of Calibration:	22-Apr-2013
PTDaemon Host System:	same as CCS
PTDaemon Host OS:	same as CCS
PTDaemon Version:	1.6.1-ddf12a8b-20130626
Setup Description:	none

Temperature Sensor temp1
Hardware Vendor:	Digi International, Inc.
Model:	Watchport/H
Driver Version:	4.20.0.0
Connectivity:	USB
PTDaemon Host System:	same as CCS
PTDaemon Host OS:	same as CCS
Setup Description:	1 cm in front of SUT main intake

Notes

None

http://www.spec.org - [email protected]

SPECpower_ssj2008 Reporter Version: [SSJ 1.2.9, July 28, 2011]