BM Seer Unofficial thoughts from an anonymous Sun employee

SAP-SD 2-Tier ECC 6.0 2-processor world record: Sun SPARC Enterprise T5240

Wednesday Apr 09, 2008

The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors set a new World Record for two processor systems achieving 4170 SD benchmark users on the two-tier SAP Sales and Distribution (SD) standard SAP ERP 2005 application benchmark.

This benchmark was run with Oracle 10g. Oracle 10g performed very well on the Sun SPARC Enterprise T5240 server.

The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors is the first two processor system to exceed 20,000 SAPS.

The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors showed a 1.9x improvement compared to Sun SPARC Enterprise T5120 with a single 1.4 GHz UltraSPARC T2 processor. The UltraSPARC T2 Plus has twice the thread count of the UltraSPARC T1.

Note:SAP does not allow vendors to make any public SWaP or price-performance comparisons to be made. The SWaP metric is a measure of server efficiency ratio that includes system performance, power and space consumption on a specific benchmark. (SWaP = Perf /[ Space (RU) x Watts ] )

The Sun SPARC Enterprise T5240 server (2RU) with a 1.4 GHz UltraSPARC T2 Plus processor outperformed the 8-core IBM System p570 (8 RU) with four 4.7 GHz POWER6 dual-core processors by 4%.

The Solaris Operating System, used in this benchmark, clearly demonstrates that the system can support over four thousand concurrent users accessing the SAP application server.

The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors outperformed the HP DL580 with four 2.93 GHz Xeon quad-core processors by 13%.

The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors outperformed all other 1-chip and 2-chip systems.

SAP-SD 2-Tier Performance Table (in decreasing performance order),all results in table are BM rev. 6.0

Benchmark Users (bigger is better)

System SAP
Users		 # CPU, GHz Mem OS DB LI/Hr SAPS Date
Sun SPARC Enterprise T5240 4170 2 UltraSPARC T2 Plus 1.4 128 GB Solaris 10 Oracle 10g 418,000 20,900 09-Apr-08
IBM p 570 4010 4 POWER6 4.7 64 GB AIX 5L
Version 5.3		 Oracle 10g 402,330 20,120 21-May-07
HP ProLiant DL580 G5 3705 4 Quad-Core Xeon 2.93 64 GB Windows Server
2003 EE		 SQL Server 2005 370,670 18,530 03-Sep-07
IBM p 550 3104 4 POWER6 4.2 64 GB Redhat Enter-
prise
Linux 5		 DB2 9.5 312,670 15,630 29-Jan-08
Fujitsu Siemens PRIMERGY TX300 S4 2335 2 QC Xeon 3.16 32 GB Windows
Server 2003 EE		 SQL Server 2005 237,000 11,850 30-Nov-07
Sun SPARC Enterprise T5120 2175 1 UltraSPARC T21.4 64 GB Solaris 10 Oracle 10g 219,000 10,950 09-Oct-07
HP ProLiant DL385 G5 2102 2 QC Opteron 2.3 32 GB Windows Server 2003 EE SQL Server 2005 210,330 10,520 28-Mar-08
HP ProLiant BL460c 2080 2 Xeon 3.0 32 GB Windows Server 2003 EE SQL Server 2005 208,000 10,400 03-Sep-07
HP ProLiant DL380 G5 2080 2 Xeon 3.0 32 GB Windows Server 2003 EE SQL Server 2005 208,670 10,430 03-Sep-07
IBM p 570 2035 2 POWER6 4.7 32 GB AIX 5L Version 5.3 Oracle 10g 203,670 10,180 21-May-07
Fujitsu Siemens PRIMERGY TX300 S4 1475 2 DC Xeon 3.33 32 GB Windows Server 2003 EE SQL Server 2005 147,670 7,380 07-Dec-07

Complete benchmark results may be found at the SAP benchmark website http://www.sap.com/benchmark.

Benchmark Description

The SAP Standard Application SD (Sales and Distribution) Benchmark is a two-tier ERP business test that is indicative of full business workloads of complete order processing and invoice processing, and demonstrates the ability to run both the application and database software on a single system. The SAP Standard Application SD Benchmark represents the critical tasks performed in real-world ERP business environments.

SAP is one of the premier world-wide ERP application providers, and maintains a suite of benchmark tests to demonstrate the performance of competitive systems on the various SAP products.

Disclosure Statement:

Two-tier SAP Standard Sales and Distribution (SD) standard SAP ERP 2005 application benchmark: SPARC Enterprise Model T5240, 2 processors / 16 cores / 128 threads, UltraSPARC T2 Plus, 1.4 GHz, 8 KB(D) + 16 KB(I) L1 cache per core, 4 MB L2 cache per processor, 128 GB main memory; Number of benchmark users & comp.: 4,170 SD (Sales & Distribution); Average dialog response time: 1.97 seconds; Throughput: Fully Processed Order Line items/hour: 418,000, Dialog steps/hour: 1,254,000; SAPS: 20,900; Average DB request time (dia/upd): 0.085 sec / 0.238 sec; CPU utilization of central server: 99%; Operating System central server: Solaris 10; RDBMS: Oracle 10g; SAP ECC Release: 6.0. The SAP certification number was not available at press time and can be found at the following Web page: www.sap.com/benchmark. SPARC Enterprise Model T5120 (1-way, 1 proc, 8 cores, 64 threads) 1 x 1.4 GHz UltraSPARC T2, 64GB memory, 2175 SD Benchmark users, 1.91 sec avg response time, Cert#2007059, Oracle 10g, Solaris 10; SPARC Enterprise Model T2000 | Sun Fire T2000 (1-way, 1 proc, 8 cores, 32 threads) 1 x 1.4 GHz UltraSPARC T1, 64GB memory, 1100 SD Benchmark users, 1.91 sec avg response time, Cert#2007051, Oracle 10g, Solaris 10; IBM System p 570 (4-way, 4 processors, 8 cores, 16 threads) 4 x 4.7 GHz POWER6, 64GB memory, 4010 SD Benchmark users, 1.96s avg resp time, Cert#2007038, Oracle 10g, AIX 5L Version 5.3; IBM System p 550 (4-way, 4 processors, 8 cores, 16 threads) 4 x 4.2 GHz POWER6, 64GB memory, 3104 SD Benchmark users, 1.91s avg resp time, Cert#2008002, DB2 9.5, Redhat Enterprise Linux 5; IBM System p 570 (2-way, 2 processors, 4 cores, 8 threads) 2 x 4.7 GHz POWER6, 32GB memory, 2035 SD Benchmark users, 1.99s avg resp time, Cert#2007037, Oracle 10g, AIX 5L Version 5.3; HP ProLiant DL580 G5 (4-way, 4 processors, 16 cores, 16 threads) 4 x 2.93 GHz Quad-Core Intel Xeon, 64GB memory, 3705 SAP SD Benchmark users, 1.99 sec avg response time, Cert#2007056, SQL Server 2005, Windows Server 2003 Enterprise Edition; HP ProLiant BL460c (2-way, 2 processors, 8 cores, 8 threads) 2 x 3.0 GHz Quad-Core Intel Xeon, 32GB memory, 2080 SAP SD Benchmark users, 1.99 sec avg response time, Cert#2007054, SQL Server 2005, Windows Server 2003 Enterprise Edition; HP ProLiant DL380 G5 (2-way, 2 processors, 4 cores, 4 threads) 2 x 3.0 GHz Quad-Core Intel Xeon, 32GB memory, 2080 SAP SD Benchmark users, 1.95 sec avg response time, Cert#2007057, SQL Server 2005, Windows Server 2003 Enterprise Edition; HP Integrity rx6600 (4-way, 4 processors, 8 cores, 16 threads) 4 x 1.6 GHz Dual-Core Intel Itanium, 48GB memory, 2150 SAP SD Benchmark users, 1.97 sec avg response time, Cert#2006083, Oracle 10g, HP-UX 11iV3;

SAP, R/3, mySAP reg TM of SAP AG in Germany and other countries. More info www.sap.com/solutions/benchmark.

Results Summary
Certified Results
Performance: 4170 benchmark users
Server: Sun SPARC Enterprise T5240
Processors: 2 x 1.4 GHz UltraSPARC T2 Plus
Memory: 128 GB
Operating system: Solaris 10
Database S/W: Oracle 10g
SAP S/W: SAP ECC 6.0
Storage: 1 x Sun StorEdge(tm) SE6120 FC Array Rack Mounted

[5] Comments
Like this post? del.icio.us | furl | slashdot | technorati | digg
Posted at 07:17AM Apr 09, 2008 in Benchmarks
Tags:
Comments:

>The Sun SPARC Enterprise T5240 server with two 1.4 GHz UltraSPARC T2 Plus processors

2xUltraSPARC T2? Is it SMP ? Cost ?

Posted by Triffids on April 09, 2008 at 08:10 AM PDT #

55000$. Thats the problem: The numbers look impressing when compared to Power, because Power is expensive. They look much weaker when compared to x64. Example: HP ProLiant DL580 G5 Rack Server 23000$, but only 10% slower....

Posted by Axel on April 09, 2008 at 08:44 AM PDT #

2Axel

1. X86 slower only on MSSQL, on equal hardware Oracle faster 20-40%. Check Oracle and MSSQL results HP Integrity rx6600 (on Oracle 30% faster) and Fujitsu PRIMEQUEST 580 (On oracle 41% faster).

2. Costly IBM p 570 on Power6 hi-end system and can scale to 8 SMP cpu's and 4 NUMA boxes and for T5240 2 cpu - maximum ...

Posted by Triffids on April 09, 2008 at 09:09 AM PDT #

Please don't comment on prices on the SAP-SD benchmark. AND when you do talk about prices on other benchmarks don't just throw out numbers without reasonable memory size and configuration details. That is the simple way to lie with X64 based system price: marketing can start with a low entry price, but a reasonable production price can be much higher.

Posted by BM Seer on April 09, 2008 at 02:32 PM PDT #

The reason is that SAP-SD does not allow $ or $/perf or watt/perf metrics to be constructed from their results. This is just part of their benchmark rules. Sun has to follow them.

Posted by BM Seer on April 09, 2008 at 03:57 PM PDT #

Post a Comment:
Comments are closed for this entry.