Skip to main content

IBM Systems  >  System p  >  News > 

New IBM server achieves championship benchmark results

Offering industry-leading performance on transaction processing, SAP, HPC and Parallel Computing, IBM System p™ servers are ideal for both commercial and technical workloads

In the key transaction processing benchmark, IBM beats HP by more than three to one

 

ARMONK, NY - July 25, 2006 - The IBM System p5™ 595 server announced today has seized the number one spot in five key industry benchmarks, demonstrating unmatched strength in both commercial and technical workloads. The dazzling performance underpins the flexibility of IBM’s Power Architecture™ technology, which serves as the foundation of systems ranging from video games to transaction processing machines to the world’s most powerful supercomputers.

The quintet of IBM benchmark victories for the 64-core System p5 follows:

Transaction Processing – In the TPC-C benchmark, measuring the ability of a server to process complex online transactions and large volumes of business data the System p5 595 running a single instance of the IBM DB2® 9 data server on the AIX 5L™ operating system and using IBM System Storage™ DS4800 processed 4,016,222 transactions per minute (tpmC) with a price/performance of $2.98/tpmC1, versus the HP Integrity Superdome’s performance of 1,231,433 tpmC at $4.82/tpmC1. The TPC-C benchmark is an industry standard for measuring the ability of a system to process complex online transactions and large volumes of business data. The TPC-C benchmark is unique in the way it exercises all components of a system, including processors, memory, networking, storage, operating system and database software, demonstrating total system performance in a way that many of the other benchmarks touted by some competitors do not.

SAP® Benchmark Result - In the two-tier SAP Sales and Distribution (SD) Standard Application Benchmark, the IBM System p5 595 running a single instance of the IBM DB2 9 data server and using IBM System Storage DS8300 achieved 23,456 Sales and Distribution Benchmark users versus Fujitsu-Siemens’ result of 21,000 Sales and Distribution Benchmark users2. The benchmark evaluates the performance of a single server running both the database and application for a Sales and Distribution environment using a standard application benchmark provided by SAP AG.

High Performance Computing – The one-core p5-595 achieved a world record result of 3,642 in the SPECfp2000 benchmark3, a key yardstick of supercomputing performance, measuring how fast a single processor, cache, and memory can run a collection of 14 floating-point compute-intensive programs.

Parallel computing – The p5-595 achieved a result of 157,880 in the SPECompMpeak20014, measuring a system's parallel processing capabilities for medium problem sizes using a suite of applications based on the OpenMP standards for shared-memory parallel processing. It is used to gauge effectiveness for high performance and technical computing.

Supersized parallel computing – The p5-595 achieved a result of 1,056,459 in the SPECompLpeak20015, which measures a system's parallel processing capabilities for large problem sizes using a suite of applications based on the OpenMP standards for shared-memory parallel processing.


IBM SPEC results were submitted to SPEC on 7/25/06.
All competitive benchmark results current as of 7/25/06.
Source of TPC-C results is http://www.tpc.org.
Source of SPEC results is http://www.spec.org.
Source of SAP results is http://www.sap.com/solutions/benchmark/index.epx

1. IBM TPC-C result of 4,016,222 tpmC, $2.98/tpmC on a 64-core (32 processor chips, 128 threads) 2.3 GHz POWER5+ System p5-595 (configuration planned to be available 12/20/06) running DB2 9.1 on AIX 5L V5.3 vs. HP TPC-C result of 1,231,433 tpmC, $4.82/tpmC on a 64-core (64 chips, 64 threads) 1.6 GHz Intel® Itanium® 2 Integrity Superdome (configuration available 6/05/06)

2. The IBM System p5 595 (2.3 GHz) ran the IBM DB2 Universal Database 9.1, AIX 5L V5.3 and the mySAP™ ERP 2004 application. Results: 1.98 seconds average response time, 2,350,330 fully processed order lines items/hours, 7,051,000 dialog steps/hour, 117,520 SAPS, 0.019 sec / 0.016 sec average database request time (dia/upd), 99% CPU utilization of central server, the SAP ECC Release 5.0 component, 64 processors, 64 cores, 128 threads. The SAP benchmark certification number was not available at press time and can be found at: http://www.sap.com/benchmark.
The Fujitsu PRIMEPOWER 2500, 128-way SMP, SPARC64 V, 2080MHz, ran the Oracle Database 9i, Solaris 9, and the SAP R/3® Enterprise Release 4.7 solution. Results: 1.91 seconds average response time, 2,116,330 fully processed order lines items/hours, 6,349,000 dialog steps/hour, 105,820 SAPS, 0.026 sec / 0.021 sec average database request time (dia/upd), 98% CPU utilization of central server, [already noted] SAP certification #2005013. Information current as of July 25, 2006.

3. IBM System p5-595 1-core 2.3 GHz (1-core, 1 thread) SPECfp2000 result of 3,642;

4. IBM System p5-595 64-core 2.3 GHz (64-core, 32-chips, 128 threads) SPECompMpeak2001 result of 157,880;

5. IBM System p5-595 64-core 2.3 GHz (64-core, 32-chips, 128 threads) SPECompLpeak2001 result of 1,056,459

IBM is a trademark of IBM Corporation in the United States and/or other countries. All other company/product names and service marks may be trademarks or registered trademarks of their respective companies. UNIX is a registered trademark in the United States and other countries licensed exclusively through The Open Group.
SPEC and the benchmark names SPECfp and SPEComp, are registered trademarks of the Standard Performance Evaluation Corporation
TPC-C is a trademark of the Transaction Processing Performance Council (TPC).



 
Contact
Jill Holt
IBM Media Relations
914.766.4123
jolt@us.ibm.com

Additional information
Benchmarks