Measuring Software Performance on Linux
Hello everyone, you can download Measuring Software Performance on Linux, a free and open source research paper provided by Martin Becker and Samarjit Chakraborty from the Technical University of Munich, which is Germany for some of you who might not know.
I have glanced through the paper, it’s not too long but it provides great information and a neat introduction on how we should setup and conduct software performance monitoring and measuring on Linux/Ubuntu systems. What I liked about this paper is that it quickly reminds us that, despite most of us getting performance results from blogs, YouTube videos and other articles, most of the time we forget to ask how exactly the person setup their infrastructure. The setup of his or her system can alter the results significantly and this is where the research paper earns my vote. With that said, go ahead and check it out and let us know what you think about it, I have included 3 formats for those of you who might prefer something other than .PDF and an abstract that tells you what to expect below the download links. Enjoy.
Table of Contents
Measuring Software Performance on Linux Research Paper in PDF
Measuring Software Performance on Linux Research Paper in Mobi
Measuring Software Performance on Linux Research Paper in EPub
What Measuring Software Performance on Linux About?
“Measuring and analyzing the performance of software has reached a high complexity, caused by more advanced processor designs and the intricate interaction between user programs, the operating system, and the processor’s microarchitecture. In this report, we summarize our experience about how performance characteristics of software should be measured when running on a Linux operating system and a modern processor. In particular, we provide a general overview about hardware and operating system features that may have a significant impact on timing and how they interact, we identify sources of errors that need to be controlled in order to obtain unbiased measurement results, and we propose a measurement setup for Linux to minimize errors. Although not the focus of this report, we describe the measurement process using hardware performance counters, which can faithfully reflect the real bottlenecks on a given processor. Our experiments confirm that our measurement setup has a large impact on the results. More surprisingly, however, they also suggest that the setup can be negligible for certain analysis methods. Furthermore, we found that our setup maintains significantly better performance under background load conditions, which means it can be used to improve software in high-performance applications.” – Abstract.