Using two of Oracle's Sun Fire X4470 servers to run the SAP Enhancement Package 4 for SAP ERP 6.0 (Unicode) Sales and Distribution ? Parallel (SD-Parallel) standard application benchmark, Oracle delivered a world record result.This was run using Oracle Solaris 10 and Oracle 11g Real Application Clusters (RAC) software.

  • The Sun Fire X4470 servers result of 21,000 users delivered more than twice the performance of the IBM System x3850 X5 system result of 10,450 users.

  • The Sun Fire X4470 servers result of 21,000 users beat the HP ProLiant DL980 G7 system resultof 18,180 users.Both solutions used 8 Intel Xeon X7560 processors.

  • The Sun Fire X4470 servers result of 21,000 users beat the Fujitsu PRIMEQUEST 1800E system resultof 16,000 users.Both solutions used 8 Intel Xeon X7560 processors.



  • This result shows how a compete software and hardware solutionfrom Oracle, using Oracle RAC, Oracle Solaris and along withOracle's Sun servers, can provide a superior performing solutionwhen compared to the competition.
Performance Landscape

SAP Enhancement Package 4 for SAP ERP 6.0 (Unicode) Sales and Distribution Benchmark, select results presented in decreasing performance order.Both Parallel and 2-Tier solution results are listedin the table.





System OS
Database Users SAPS Type Date Two Sun Fire X4470
4xIntel Xeon X7560 @2.26GHz
256 GB Solaris 10
Oracle 11g Real Application Clusters 21,000 115,300 Parallel 28-Jun-10 HP DL980 G7
8xIntel Xeon X7560 @2.26GHz
512 GB Win Server 2008 R2 DE
SQL Server 2008 18,180 99,320 2-Tier 21-Jun-10 Fujitsu PRIMEQUEST 1800E
8xIntel Xeon X7560 @2.26GHz
512 GB Win Server 2008 R2 DE
SQL Server 2008 16,000 87,550 2-Tier 30-Mar-10 Four Sun Blade X6270
2xIntel Xeon X5570 @2.93GHz
48 GB Solaris 10
Oracle 10g Real Application Clusters 13,718 75,762 Parallel 12-Oct-09 IBM System x3850 X5
4xIntel Xeon X7560 @2.26GHz
256 GB Win Server 2008 EE
DB2 9.7 10,450 57,120 2-Tier 30-Mar-10 HP DL580 G7
4xIntel Xeon X7560 @2.26GHz
256 GB Win Server 2008 R2 DE
SQL Server 2008 10,445 57,020 2-Tier 21-Jun-10 Fujitsu PRIMERGY RX600 S5
4xIntel Xeon X7560 @2.26GHz
512 GB Win Server 2008 R2 DE
SQL Server 2008 9,560 52,300 2-Tier 06-May-10 Two Sun Blade X6270
2xIntel Xeon X5570 @2.93GHz
48 GB Solaris 10
Oracle 10g Real Application Clusters 7,220 39,420 Parallel 12-Oct-09 One Sun Blade X6270
2xIntel Xeon X5570 @2.93GHz
48 GB Solaris 10
Oracle 10g Real Application Clusters 3,800 20,750 Parallel 12-Oct-09 Sun Fire X4270
2xIntel Xeon X5570 @2.93GHz
48 GB Solaris 10
Oracle 10g 3,800 21,000 2-Tier 21-Aug-09 Complete benchmark results may be found at the SAP benchmark websitehttp://www.sap.com/solutions/benchmark/sd.epx.

Results and Configuration Summary

Hardware Configuration:


2 x Sun Fire X4470 servers, each with
4 x Intel Xeon X7560 2.26 GHz (4 chips, 32 cores, 64 threads)
256 GB memory

Software Configuration:


Oracle 11g Real Application Clusters (RAC)
Oracle Solaris 10
Results Summary:


Number of SAP SD benchmark users:
21,000 Average dialog response time:
0.93 seconds Throughput:

Dialog steps/hour:
6,918,000
SAPS:
115,300 SAP Certification:
2010029
Benchmark Description

The SAP Standard Application Sales and Distribution - Parallel (SD-Parallel) 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.

The SD-Parallel Benchmark consists of the same transactions and userinteraction steps as the SD Benchmark. This means that the SD-ParallelBenchmark runs the same business processes as the SD Benchmark. Thedifference between the benchmarks is the technical data distribution.

An additional rule for parallel and distributed databasesis one must equally distribute the benchmark users across all databasenodes for the used benchmark clients (round-robin-method). Followingthis rule, all database nodes work on data of all clients. This avoidsunrealistic configurations such as having only one client per databasenode.

The SAP Benchmark Council agreed to give the parallel benchmark adifferent name so that the difference can be easily recognized by anyinterested parties - customers, prospects, and analysts. The namingconvention is SD-Parallel for Sales & Distribution - Parallel.

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

See Also

Disclosure Statement

SAP Enhancement Package 4 for SAP ERP 6.0 (Unicode) Sales and Distribution Benchmark, results as of 6/22/2010.For more details, see http://www.sap.com/benchmark.SD-Parallel, Two Sun Fire X4470 (each 4 processors, 32 cores, 64 threads) 21,000 SAP SD Users, Cert# 2010029.SD 2-Tier, HP ProLiant DL980 G7 (8 processors, 64 cores, 128 threads)18,180 SAP SD Users, Cert# 2010028.SD 2-Tier, Fujitsu PRIMEQUEST 1800E (8 processors, 64 cores, 128 threads) 16,00o SAP SD Users, Cert# 2010010.SD-Parallel, Four Sun Blade X6270 (each 2 processors, 8 cores, 16 threads) 13,718 SAP SD Users, Cert# 2009041.SD 2-Tier, IBM System x3850 X5 (4 processors, 32 cores, 64 threads) 10,450 SAP SD Users, Cert# 2010012.SD 2-Tier, Fujitsu PRIMERGY RX600 S5 (4 processors, 32 cores, 64 threads) 9,560 SAP SD Users, Cert# 2010017.SD-Parallel, Two Sun Blade X6270 (each 2 processors, 8 cores, 16 threads) 7,220 SAP SD Users, Cert# 2009040.SD-Parallel, Sun Blade X6270 (2 processors, 8 cores, 16 threads) 3,800 SAP SD Users, Cert# 2009039.SD 2-Tier, Sun Fire X4270 (2 processors, 8 cores, 16 threads) 3,800 SAP SD Users, Cert# 2009033.



Read More about [Sun Fire X4470 2-Node Configuration Sets World Record for SAP SD-Parallel Benchmark...