IBM’s Relative Performance (rPerf) metric is a great tool for comparing commercial workload performance between different Power System servers. It’s often. This post outlines a recent Nigel Griffiths tweet about estimating rperf for your LPAR. Read on to find out more. Articles on IBM AIX performance including server throughput, system performance and IBM AIX commands.

Author: Nibar Zolomi
Country: Honduras
Language: English (Spanish)
Genre: Finance
Published (Last): 26 November 2012
Pages: 37
PDF File Size: 7.84 Mb
ePub File Size: 7.40 Mb
ISBN: 854-6-55376-840-5
Downloads: 58315
Price: Free* [*Free Regsitration Required]
Uploader: Akibar

Recent AIX version are much better out of the box but the basics still need to be monitored and checked. Particularly as you move to larger servers, the topology of the cores and memory that are used can have a big effect on the cache and memory affinity of the LPAR and thus have a big effect on the actual performance.

IBM Systems Magazine – AIXchange blog

The amount of rPerfs required to run existing workloads can be determined and used to size new servers. How much of a reduction? Make sure you have the latest rperf version Run: But I hope this blog entry will help every one to either get it right first time or at least set realistic expectation and outline some prime areas that will need investigation to realise benefits, thanks Nigel.

Previous Entry Main Next Entry. So lets talk about things you should think about when alx rPerf to make sizing or performance estimates.

Second, rPerf numbers are expressed in terms of throughput as opposed to speed. Running the same computer benchmark on multiple computers allows for comparison. How will the script get aic

In this case, the script guesses the rPerf based on rPerf numbers in a fairly crude way. There is no way to convert a roltp number to a rPerf.

The main reason was to publish data for a broad selection gperf Power Systems servers. We use cookies to optimize your visit to our website. Numerous benchmark programs are in use today. The pSeries is the baseline reference system and has a value of 1.


This has resulted in the publication of multiple rPerf values for each server based on varying processor core counts. Running older software from the older machines can cause a mismatch with your rPerf expectation.

Bug Free – users have to be willing to upgrade firmware, AIX and application software, as necessary.

How to Use rPerfs for Workload Migration and Server Consolidation

I have done this sort of exercise myself in the past. Note that rPerf benchmark results aren’t intended to represent any specific public-benchmark results and inferences shouldn’t be made. Many benchmarks exist for comparing servers, so why did IBM create the rPerf measurement?

Note, although rPerf may be used to compare estimated IBM UNIX commercial processing performance, actual system performance may vary and is dependent upon many factors, including system hardware configuration, software design, and configuration. Since IBM owns and controls the rPerf benchmark, it can report both actual and estimated data for Power servers.

Firmware is Current – The firmware includes the Hypervisor and this has many performance enhancement tweaks and often based on field experience that you need working for you. These guidelines may help you achieve the performance that you expected.

If you would like to change your cookie settings at any time please view our privacy policy for additional information.

Continue to Wikis

Downtime needs to be built-in to allow the removal of problems in the above. This is normal tuning like disk queue depth, memory use, large network packets.

Performance tuning is always an iterative process of removing one performance bottleneck in order to reveal the next performance bottleneck. This is the configuration used in working out rPerfs and unused DIMM slots or minimum memory will reduce performance.

Having this comprehensive set of information makes it easier to estimate the commercial workload CPU performance for any Power Systems server. Actual performance will vary, based on application and configuration details. The biggest pitfall is trying to use benchmarks without knowing the system bottleneck.


Check the Facts and Features document for the rPerf with smallest number of CPU cores for you machine, below that number you are making assumptions – particularly as you go below boundaries in the machine like a drawer or whole POWER7 chip and below one CPU core.

These requirements would make it extremely costly to publish multiple data points rperr each Power Systems model. Note that aiix can only be used for making comparisons within Power servers. He can be reached through www. He can be reached at ccler forsythe.

These are a simple calculation and will not be exact – i. Also no SAN bottlenecks. The most exciting POWER6 enhancement, live partition mobility, allows one to migrate a running LPAR to another physical box and is designed to move running partitions from one POWER6 processor-based server to another without any application downtime whatsoever.

The rPerf Benchmark

If you are breaking a lot of these assumptions then you may not get the rperff you expected or hoped to achieve. We use cookies to optimize your visit to our website. Specifically, a computer benchmark is a computer program that performs a strictly defined workload set of operations and returns some form of measurable result i. I call these my Ten Golden Rules or considerations to remember while using rPerfs for sizing. The official rPerf numbers for current and older machines can be found here: If you are some where between these options then you may have to make a series of changes, upgrades or updates to achieve good performance on your new machine.