Using Microsoft’s Performance Toolkit for Benchmarking – Part 2

Part I of this blog series discussed the benefits of using the Windows Assessment & Deployment Toolkit for benchmark testing. This blog will provide further detail around how to install the required component of the Windows Assessment and Deployment Kit.

When installing the Windows Assessment and Deployment Kit, the fourth screen of the installation process, depicted in Figure 1, allows the user to choose which component to install. The only component required for benchmarking is the “Windows Performance Toolkit”, select that component and proceed with the installation.

Figure 1: A screenshot of the Windows Performance Toolkit where the user selects “Windows Performance Toolkit”.


After the installation is complete, the next step is to launch “Windows Performance Recorder”.  This can be accomplished several ways, one method is to simply type “Windows Performance” in the search box after clicking on the Windows Start button.  The GUI pictured below is available to execute tests, however this process can also be automated by using the command line feature of Performance Recorder.

Figure 2: A screenshot of the “Windows Performance Recorder” GUI.

An important step in using this tool is to determine how many tests should be run in order to obtain valid results. A typical project may have a base system image, with multiple configurations.In order to obtain better statistical measurements, Olenick recommends running a 24hr test on each platform’s base configuration, and 20 iteration tests for modifications to that system, figure 3 depicts these test settings.

Figure 3: A screenshot depicting the recommended benchmarking tests to execute.

Stay tuned for part three of this blog series where we will dive deeper into the data analysis.

Author: Mike Willett, Lead senior consultant at Olenick    

Connect with Olenick
 image spacerFollow us on Twitterimage spacerOlenick LinkedIn Pageimage spacerLike us on Facebookimage spacerG+image spacer