Results of the PeopleSoft Campus Solutions 9.0 benchmark on Sun SPARC Enterprise M4000 and X6270 blade servers are available. This benchmarking report is particularly significant, writes Giri Mandalika in his Scratchpad blog, because the workload has both online transactions and batch processes and, furthermore, it is the first time Sun has published a PeopleSoft benchmark on x64 hardware running Oracle Enterprise Linux.

Results are posted in terms of hourly throughput (invoices & transcripts per hour). The published benchmark exercise is described as measuring the comparative online and batch performance of selected processses in Oracle's PeopleSoft Enterprise Campus Solutions 9 with Oracle 11g aa.1.0.7. The test used a single 2-way Quad-Core (8 cores in all) Sun SPARC Enterprise M4000 Server, running Solaris 10 10/09 for the database server. Two 2-way (16 cores total) Sun Blade X6270 Server Modules were used as Application Servers running Oracle enterprise Linux 4 Update 8, and a single 2-way (8 cores total) X6270 was used as the Web Server and Process Scheduler.

Pitted against two HP servers -- an rx6600 and a ProLiant DL580 -- the Sun solutions delivered batch throughput/hour rates of 31,797 (invoices) and 36,652 (transcripts) vs. 22,753 and 30,257 for the rx6600 and 17,621 and 25,423 for the ProLiant DL580.

Mandalika notes that response times for the online transactions were measured using HP's QuickTest Pro (QTP) tool whose test scripts have a dependency on the web browser (IE in particular) -- hence it is extremely sensitive to the web browser latencies, remote desktop/VNC latencies and other latencies induced by the operating system, he cautions readers. All these latencies will be factored into the transaction response times and, as a result, the final average transaction response times might be skewed a little, he writes.

The blog also points out that the PeopleSoft Process Scheduler (batch server) was set up on the same node as that of the web server node. In general, Oracle recommends setting up the process scheduler either on the database server node or on a dedicated system. In the benchmark environment, the decision was made not to run the process scheduler on the database server node as it would hurt the performance of the online transactions, Mandalika writes. At the same time, there were plenty of idle CPU cycles on the web server node even at the peak load of 6,000 concurrent users, so the decision was made to run the PS on the web server node. He suggests that, in case customers are not comfortable with this kind of setup, they can use any supported virtualization technology (eg., Logical Domains, Containers on Solaris, Oracle VM on OEL) to separate the process scheduler from the web server by allocating the system resources as they like.

Mandalika also mentions that the load balancing mechanism in PeopleSoft forwards the incoming requests to the appropriate application server in the enterprise, and within the application server, sends the request to an appropriate application server process, PSAPPSRV.

Another food-for-thought note is that the application server balances the load among application server processes in a round robin fashion on *nix platforms whereas on Windows, it forwards all the requests to a single application server process until it reaches the configured limit before moving on to the next available application server process. Most of the time, the number of requests processed by each of the identically configured application server processes [running on different application server nodes in the enterprise] may not be even, which could lead to uneven resource usage across different nodes in the PeopleSoft deployment. This is apparent in the CPU and memory usage reported for the two app server nodes in the benchmark environment, Mandalika writes.

Additionally, Mandalika notes that I/O latency, hot tables and hot indexes were placed on a Sun Flash Accelerator F20 PCIe Card in this benchmark. The F20 card has a total capacity of 96 GB with 4 x 24GB Flash Modules (FMODs). Although this workload is moderately I/O intensive, the batch processes in this benchmark generate a lot of I/O for few minutes in the steady state of the benchmark. The flash accelerator handled the burst of I/O activity pretty well, and as a result the performance of the batch processing was improved.

The author adds that no tuning was performed on the Solaris 10 OS used in this benchmark.

More Information

PeopleSoft Campus Solutions 9.0 benchmark on Sun SPARC Enterprise M4000 and X6270 blades - Oracle/PeopleSoft/Sun Benchmark results

PeopleSoft Campus Solutions 9.0 benchmark on Sun SPARC Enterprise M4000 and X6270 blades - Mandalika's blog entry

Comparative Table of Benchmark Test Results - from Mandalika

Deploying Oracle PeopleSoft Campus Solution on Sun Storage 7000 Unified Storage Systems

World Record Performance on PeopleSoft Payroll (North America) 9.0 Benchmark [...read more...]

		fullsource
Print Email Current Profile Archive Blog Generate newsletter