World Record SPEC OMP2001 Sun Blade X8420

Thursday Jan 11, 2007

The Sun Fire X8420 M2 topped the IBM System p5 550 (2.1 GHz, 4 CPU) result by 16%. The Sun Blade X8420 delivers the best performance on the SPEC OMPM2001 benchmark suite of all systems running 8-threads results using dual-core AMD Opteron 8220, produced best SPECompM2001 result of 23224.

The results show that the combination of Solaris 10 using Sun Studio 11 is unmatched by the competition for assisting users in writing parallel code.

More info on sun.com X8420 Benchmark Page.

SPECompM2001 (bigger is better, ordered by peak metric)

Result		Cores	Chips	Thrds	System
Peak	Base
23224	22531	8	4	8	Sun Blade X8420 M2, Opteron 8220, 2.8GHz,64GB
21167	20409	8	4	8	Sun Fire X4600 M2, Opteron 8220SE, 2.8GHz,16GB
20319	19708	8	8	8	Sun Fire X4600, Opteron 856, 3.0GHz
19983	15355	4	2	8	IBM System p5 550 (2.1 GHz, 4 CPU)
19653	18949	8	4	8	Sun Blade X8400, Opteron 885, 2.6GHz
17948	16621	8	4	8	HP DL585, Opteron 880, 2.4GHz
16096	14335	8	4	8	IBM 570, POWER5, 1.9 GHz

Benchmark Description

The SPEC OMPM2001 Benchmark Suite was released in June 2001 and tests HPC performance using OpenMP for parallelism.

• 11 programs (3 in C and 8 in Fortran) parallelized using OpenMP API

Goals of suite:

• Targeted to mid-range (4-32 processor) parallel systems
• Run rules, tools and reporting similar to SPEC CPU2000
• Programs representative of HPC and Scientific Applications

Disclosure Statement:

SPEC, SPEComp reg tm of Standard Performance Evaluation Corporation. Results from www.SPEC.org as of Jan 08, 2007, Sun result submitted to SPEC. Sun Blade X8420 (8 cores, 4 chips, 8 threads), 23,224 SPECompM2001. IBM System p5 550 (4 cores, 2 chips, 8 threads), 19,983 SPECompM2001. Sockets refers to chips.

Results Summary

Result
	X8420 8-threads:		23224 SPECompM2001
Reference Date:			Jan 09, 2007
System:			Sun Blade X8420
Total Memory :			64 GB DDR667 (16x4GB DIMMs)
Processors/GHz:			Four Opteron 8220, 2.8 GHz
Operating System:			Solaris 10
Compiler:			Sun Studio 11

1.4 GHz Sun Fire T2000 Lotus World Record per-socket performance and Perf/watt

Wednesday Jan 10, 2007

Lotus Domino NotesBench R6iNotes Sun Fire T2000 Lotus World Record per-socket performance and Perf/watt

The Sun FireT2000 1.4GHz US T1 using Lotus Domino 7.0.2 mail server delivered world record results with best per-socket performance of 23200 and world record performance/watt for the Lotus[R] R6iNotes on Domino mail HTTP based R6iNotes benchmark.

• The Sun Fire T2000 is 22% faster with 25% faster response times than the latest HP DL380G5 based on two 3GHz dual-core Intel 5160 (WoodCrest) running Microsoft Windows Server 2003. This translates to 2.4x higher performance per socket.

• The Sun Fire T2000 delivers this performance while consuming nearly three times less power, adding up to 3.4x higher Performance per Watt and SWaP. While the HP system achieved lower price/performance, for many customers, this would be offset by the higher energy consumption of the system.

• When compared to IBM's latest p5+ 550Q server, based on 2 x Quad Core p5+ processors and the AIX operating system, the Sun Fire T2000 was able to deliver comparable performance with 35% faster response times, and 2x higher Performance per Socket.

• Also, the Sun Fire T2000 Server consumed 2.5x less power, and only half the space of the IBM system, adding up to 5X higher SWaP.

• In addition to the power, cooling and space savings, the Sun Fire T2000 also delivered one third superior price / performance than IBM system.

  Lotus Domino 7.0 NotesBench R6iNotes Performance Chart (in decreasing #Users order)

  System	$/User	Users	NotesMark	#Partitions	Response
  Sun Fire V890 8xUltraSPARC IV+ 1800MHz Solaris10	$7.19	40000	33862	4	324 ms
  IBM eServer Power5 550Q 8x1.5GHz POWER5, AIX 5L V5.3	$5.97	24000	20108	4	932 ms
  Sun Fire T2000 1xUltraSPARC T1 1400MHz Solaris 10	$4.48	23200	19518	4	692 ms
  IBM X3650, 4x3.0GHz Xeon SUSE 9	$3.47	22000	18989	4	>3 Sec
  Sun Fire T2000 1xUltraSPARC T1 1200MHz Solaris 10HW2	$3.94	19000	16061	4	400 ms
  HP DL380G5, 2x3.0GHz Xeon 5160, Windows Server 2003	$3.15	19000	15750	3	868 ms
  HP Proliant DL580 G3 4x3.0GHz Xeon, Windows Server 2003	$4.29	18500	15953	4	434 ms

  Complete benchmark results may be found at the Lotus NotesBench website http://www.notesbench.org. Benchmark version is Notesbench 6.5 running on Lotus Domino 6.5.1

Benchmark Description

The benchmark simulates active users accessing their Domino[R] R6iNotes mail files via standard Web browser. Each simulated user periodically sends, retrieves, and deletes a specified number of e-mail messages from a browser. An average user runs this script four times per hour.

The R5iNotes and R6iNotes workloads, using the Lotus Domino Mail server (R5 or R6) are both HTTP based workloads. R6iNotes is heavier with added features and larger mail files using the MIME format.

The Lotus Webmail and iNotes workloads are NOT comparable.

see also:

NotesBench Results on Ideas International

NotesBench Website

Disclosure Statement:

NotesBench R6iNotes Sun Fire T2000(1chip, 8cores/chip@1.4GHz UltraSPARC T1, 4threads/core, 64GB), 4 partitions, Solaris[TM] 10 U3, Lotus[R] Domino 7.0.2, 23200 users, $4.48per user, 19518 NotesMark tpm, 692ms avg rt. , IBM eServer 550Q, 8x1.5GHz POWER5, 32GB, 4 partitions, AIX 5L V5.3, Lotus Domino 7.0, 24000 users, $5.97 per user, 20108 NotesMark tpm, 932 ms avg rt., HP DL380G5, 2x3.0GHz, Intel Xeon 5160, 6GB, 3 partitions, Windows Server 2003, Lotus[R] Domino 7.0.1, 19000 users, $3.15 per user, 15750 NotesMark tpm, 868ms avg rt., More info: www.notesbench.org

HP DL380G5 power consumption estimated by taking 70% of the maximum input wattage rating of 1193W reported here on 11/14/06: http://h18004.www1.hp.com/products/quickspecs/12477_na/12477_na.html#Power%20Specifications IBM p5 550Q power specifications calculated by applying 70% of the maximum Watts published in “Facts and Features Report”, 11/14/06, posted at: ftp:/ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF

SunFire T2000 1.4GHz server Power consumption has been taken during the full benchmark execution.

Results Summary

Audited Results
	Users:		23200
	NotesMark:		19518
	Price Performance:		$4.48 $/User
	Price Performance:		$5.33 $ /NotesMark
	Response:		692 ms
Systems:			One Sun Fire T2000
Number Processors:			1 chip, 8cores/chip@1.4GHz, 4threads/core
Speed of Processors:			1400 MHz
Storage:			3 x SS3320 (12x73GB)
Notes Version:			Lotus Domino 7.0.2
	#Domino Partition		4
Operating System:			Solaris 10 U3
Cost:			$104,051.75.00
Other Performance Metrics
	Users/CPU:		23200
	Users/Core		2900

[1] Comments

• Just wanted to commend you on your blog. These pos...
• Post a comment

Like this post? | | | |

Posted at 06:06AM Jan 10, 2007 by BM Seer in Benchmarks
Tags: fire lotus performance record sun swap t2000

NEW 1.4GHz Sun Fire T2000 sets World Record 2-Node SPECjAppServer

Tuesday Jan 09, 2007

The new 1.4GHz Sun Fire T2000 achieved 801.70 SPECjAppServer2004 JOPS and beats the HP Itanium-2 rx3600 two-node result of 618.22 SPECjAppServer2004 JOPS by 29%. This is the dual-core 1.6GHz Itanium2!

The T2000 result also beats the IBM two-node result using the p5 505Q and p5 550 of 618.38 SPECjAppServer2004 JOPS@Standard by 29%.

One Sun Fire T2000 server equipped with one 1.4GHz UltraSPARC T1 running BEA Weblogic 9.2 Advantage Edition and one Sun Fire T2000 equipped with a UltraSPARC T1 processor at 1.0 GHz running IBM DB2 8.2.6 delivered a result of 801.70 JOPS@Standard for best performance of single socket servers in the SPECjAppServer2004 benchmark.

IBM POWER5+ and Itanium2 Dual-cores also take a lot more power and space:

When compared to HP's rx4640 equipped with Itanium2 and running Linux, the T2000 delivers nearly 1.5X higher performance in 4x less power and half the space, delivering over 6X higher performance per watt and 12x SWaP.

The Sun Fire T2000 delivers nearly 1.3X higher performance in nearly 3x less power and half the space, resulting in nearly 3.5X higher performance per watt and 7x SWaP when compared to HP's rx3600 equipped with two of the latest dual core Itanium2 processors and running HP-UX11i.

This result highlights the performance benefits of the latest BEA Weblogic Server 9.2 release on Sun Fire servers. It also shows the best way to get superior performance on IBM DB2 software is to use the Sun T2000 server.

This benchmark result demonstrates that the Sun Fire T2000 running the Solaris 10 Operating system can support thousands of concurrent users accessing Web Services applications.

SPECjAppServer2004 Performance Chart - JOPS@Standard
c/c = cores/chip.

	JOPS@ Standard	J2EE Server	App SW	DB Server	DB SW
Sun	801.70	1 x Sun Fire T2000 1.4GHz US-T1 8 core/1 chip (8 c/c)	BEA WebLogic 9.2	1 x Sun Fire T2000 1.0GHz US-T1 6 core/1 chips (6 c/c)	IBM DB2 8.2.6
IBM	618.38	1 x IBM p505Q 1.65GHz POWER5+ 4 core/2 chips (2 c/c)	IBM WebSphere 6.1	1 x IBM p550 2.1GHz POWER5+ 4 cores/2 chips (2 c/c)	IBM DB2 8.2
HP	618.22	1 x HP rx3600 1.6GHz Itanium2 4 core/2 chip (2 c/c)	BEA WebLogic 9.2	1 x HP rx4640 1.6GHz Itanium2 4 core/4 chip (1 c/c)	Oracle 10g
Sun	615.64	1 x Sun Fire T2000 1.2GHz US-T1 8 core/1 chip (8 c/c)	BEA WebLogic 9.0	1 x Sun Fire V490 1.5GHz US-IV+ 8 core/4 chips (2 c/c)	Oracle 10g 10.1.0.4
HP	542.17	1 x rx4640 1.6GHz Itanium2 4 core/4 chip (1 c/c)	BEA WebLogic 9.1	1 x rx4640 1.6GHz Itanium2 4 core/4 chip (1 c/c)	Oracle 10g 10.1.0.4
IBM	404.88	1 x p5 505 2.1GHz POWER5+ 2 cores/1 chip (2 c/c)	IBM WebSphere V6.1	1 x p4 505Q 1.65GHz POWER5+ 4 core/2 chips (2 c/c)	IBM DB2 v8.2

SPECjAppServer2004 Results Page

Benchmark Description

SPECjAppServer2004 (Java Application Server) is a multi-tier benchmark for measuring the performance of Java 2 Enterprise Edition (J2EE) technology-based application servers. SPECjAppServer2004 is an end-to-end application which exercises all major J2EE technologies implemented by compliant application servers as follows:

• The web container, including servlets and JSPs
• The EJB container
• EJB2.0 Container Managed Persistence
• JMS and Message Driven Beans
• Transaction management
• Database connectivity

Moreover, SPECjAppServer2004 also heavily exercises all parts of the underlying infrastructure that make up the application environment, including hardware, JVM software, database software, JDBC drivers, and the system network.

The primary metric of the SPECjAppServer2004 benchmark is jAppServer Operations Per Second (JOPS) which is calculated by adding the metrics of the Dealership Management Application in the Dealer Domain and the Manufacturing Application in the Manufacturing Domain. There is NO price/performance metric in this benchmark.

Disclosure Statement:

SPECjAppServer2004 Sun Fire T2000 (8 cores, 1 chip) 801.70 JOPS@Standard. SPECjAppServer2004 IBM p5 505Q (4 cores, 2 chips) 618.38 JOPS@Standard. SPECjAppServer2004 HP rx4600 (4 cores, 4 chip) 542.18 JOPS@Standard. SPECjAppServer2004 IBM p5 505 (2 cores, 1chip) 404.88 JOPS@Standard. SPECjAppServer2004 HP rx3600 (4 cores, 2 chips) 618.22 JOPS@Standard.
SPEC, SPECjAppServer reg tm of Standard Performance Evaluation Corporation. All results from www.spec.org as of 01/09/07.
HP rx4640 server specifications 10/19/05 from http://www.hp.com/products1/servers/integrity/entry_level/rx4640/
HP rx4640 power rating of 1,303 watts taken from HP Enterprise Configurator 10/19/05 from http://h30099.www3.hp.com/configurator/catalog-hpintegrity.asp. System configured with Redundant Power, 4 x 1.6GHz Itanium processors, 8 x 2GB DIMMs, 0 x PCI cards and 2 x 73GB HDDs.
IBM specifications from Fact and Features report, 1/9/06: ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/PSB01628USEN/PSB01628USEN.PDF. IBM power is based on the reported maximum power consumption.
HP rx3600 power consumption estimated by taking 70% of the maximum output power supply rating reported here on 11/14/06: http://h20341.www2.hp.com/integrity/cache/387834-0-0-225-121.html
Sun Fire T2000 server power consumption taken from measurements made during the benchmark run.

Certified Result in a two-system configuration:

• Certified Result: 801.70 JOPS@Standard
• Reference Date: Jan 9, 2007

Application Server:

• Sun Fire T2000:
• one 1.4 GHz 8-core UltraSPARC T1
• 64 GB memory (16x4GB)
• Solaris 10 11/06
• BEA WebLogic 9.2 Advantage Edition
• JVM J2SE 5.0 Update 10

Database Server:

• Sun Fire T2000:
• one 1.0 GHz 6-core UltraSPARC T1 processor
• 8 GB memory
• 2x Sun StorEdge SE3320 SCSI Array
• Solaris 10 6/06
• IBM DB2 Universal Database v8.2.6

New World Record SPECint_rate2006 Sun Ultra 40 M2 Workstation

Thursday Nov 16, 2006

The Sun Ultra 40 M2 Workstation demonstrates a new World Record integer throughput performance for all x86 systems, sets a new world record on the new and improved SPEC cpu benchmark called "SPECint_rate2006." It fixes things like SPECint_rate2000 has/had floating-point applications in the integer suite, whaaaat? yes strange but true.

The Sun Ultra 40 M2 delivered the SPECint_rate2006 score of 48.8, using Solaris 10 and Studio 11 combination. Sun's Opteron beats Woodcrest by 7%. As you can see below 'Peak' means you add a few more compiler flags. I guess Woodcrest didn't have any others to try on Woodcrest or maybe they saw no improvement so they avoided publishing? Anyone know?

Competitive Landscape

Selected SPEC CPU2006 (SPECint_rate2006) Performance Results - bigger is better, see www.spec.org for complete results.

System	Processors					Performance Results
	Type	GHz	Chips	Cores	Threads	Peak	Base
Sun Ultra 40 M2	AMD Opteron 2220SE	2.8	2	4	4	48.4	41.9
HP DL585	Opteron 854	2.8	4	4	4	46.9	41.4
Supermicro X7DBE	Woodcrest, Xeon 5160	3.0	2	4	4	---	45.2
Sun Fire X4200	Opteron 285	2.6	2	4	4	42.8	37.8
Fujjitsu RX220	Opteron 280	2.4	2	4	4	40.0	35.7
Sun Fire X4200	Opteron 256	3.0	2	2	2	26.4	23.1
HP DL585	Opteron 854	2.8	2	2	2	25.2	22.3
Dell PrecWork 380	Pentium EE	3.73	1	2	2	--	23.1
HP DL380 G4	Pentium 4	3.8	2	2	2	--	20.9

Benchmark Description

SPEC CPU2006 is made up of two suites of benchmarks, CFP2006 and CINT2006. CFP2006 targets floating-point performance, while CINT2006 targets integer performance.

Each suite has two different measures. First is the CPU measure, which is the performance on the suite as a single stream. This can be either a single thread or automatic compiled parallel run. This measure is further defined by base and optimized runs. Base uses the same compiler flags for all kernels, where optimized is allowed to use different compiler flags for each kernel. Results are compared against a baseline system run that was standardized by www.spec.org.

The second measure is Rate. I think this one is a LOT more important. It is a measure of how many CPU measures can be run at a time. Typically, it is run as n processes on n processors or threads. It shows how well the same job mix can run on a system under some load. It also is run as a base and optimized set of results. "Rate" is what you use for any mult-threaded workstation and all servers.

Disclosure Statement:

SPEC, SPECint reg tm of Standard Performance Evaluation Corporation. Results from www.spec.org as of 11/14/06. Sun Ultra 40 M2, 48.8 SPECint_rate2006.

System Configuration

• Sun Ultra 40 M2
• 2 x 2.8 GHz Opteron 2220SE
• 16GB memory
• Solaris 10
• Sun Studio 11
• 48.8 SPECint_rate2006