Simple measurements of software trace data using Sourcery Analyzer

My last post covered how to import data into Sourcery Analyzer that didn’t come from the Linux Trace Toolkit. In this post I’ll show you a few simple tools you can use to inspect the data in more detail. If you are using Sourcery Analyzer its because you want to explore the software behavior or performance of your design. This task is really three parts in one:

  1. Data collection: Gathering up information from the target. Sourcery Analyzer helps by providing automated data collection for Linux and Nucleus systems and a means to import data of any type that can be expressed in a text file.
  2. Data analysis: Turning this raw data into useful answers. Sourcery Analyzer has a pre-stocked library of analysis routines or you can add your own if need be.
  3. Making comparisons and form conclusions: This is the part that is up to you and the spongy grey matter you’ve got! Sourcery Analyzer does have some tools to help, though.
Sourcery Analyzer helps you to make comparisons and draw form conclusions by creating data visualizations for you. My last post ended with this simple graph of my Honda Accord’s RPM over time:
Graph of imported data showing RPM
Graphs like this are great for quickly grasping the nature of the data. Making a comparison requires more detail, though. One of the most fundamental tools in Sourcery Analyzer is the “cursor”. This short video will demonstrate how to use cursors (yes, more than one!) to make precise observations in the graph.

One of the fundamental principles of data visualization is that a visualization needs to both show the data and make comparisons. For simple comparisons cursors can be a simple way to make your point. Often showing just a bit of data can be valuable as well. If you need to review the source data you can always do so using the Trace Viewer:
This is far more than just a simple viewer. The Trace Viewer can also display data synchronized with the graphs so that both forms can complement each other. To do so Sourcery Analyzer uses a “synchronization marker”. Synchronization markers designate a point in time of a graph or trace viewer at as being locked together. Then any scrolling that happens in one view is reflected in the other such that the same offset from the synchronization marker is displayed. In the simple case setting the synchronization marker to the beginning of the data is what most people want. This video will show you synchronization markers in action:

In the next post I’ll introduce you to the advanced measurements which can be accomplished using the other measurement tools offered by Sourcery Analyzer. If you would like to try some of this on your own please consider picking up an evaluation copy of Sourcery Analyzer. If you have questions feel free to leave a comment or drop into the Sourcery Tools forum and ask.

If you import something interesting to Sourcery Analyzer please come back and leave a comment!

Post Author

Posted May 22nd, 2013, by

Post Tags

Post Comments

1 Comment

About Embedded Blog

Open source experts discuss challenges and community happenings for embedded Linux, automotive, IVI, HMI and technology integration. Embedded Blog

Comments

One comment on this post | ↓ Add Your Own

[...] Simple measurements of software trace data using Sourcery Analyzer Why do I bring this up? I’m excited to share with you that the new 2013.05 release of Sourcery CodeBench now includes the most significant parts of Sourcery Analyzer! Any customer who has or will purchase Sourcery CodeBench Professional can take advantage of these key features of Sourcery Analyzer: [...]

Add Your Comment

Archives