In this blog post, we will see how Tracealyzer can be used to quickly and efficiently evaluate multiple implementations of an algorithm in Python, a language that is becoming more common in embedded application development as most machine learning frameworks are implemented in Python.
In this post, we’re going to understand how the combination of LTTng and Tracealyzer can shine light on how compiler options impact performance. The method discussed can come in handy whenever we are evaluating the performance of multiple candidate implementations of a particular feature.
In this blog post, we’re going to demonstrate how to create LTTng tracepoints and how to use Tracealyzer for Linux to measure certain metrics based on these tracepoints.
When developing an application that’s targeting a Linux-based system, it is important to configure our system to maximize performance, because misconfiguration can limit application performance.
Today’s post dives into a key component of a Linux device driver, the interrupt handler, and shows how Tracealyzer can give you feedback on the performance of your handler.
While there are mechanisms native to the Linux kernel to ensure that the functionality of a custom-written driver is correct, evaluating performance is not straightforward. Here’s how you can use Tracealyzer for Linux to assess the performance and identify any deficiencies.