Linux has a well-know experimental-based research approach composed of the following steps: i) implementation of your solution; ii) test and verification of your solution; iii) data collection and statistics calculation; iv) ploting of graphics; v) writing and presentation of your reports and papers. Linux has a standard set of tools that provides all you need to apply this approach. In this topic I will show you all the standard tools for each step of the approach and point the best tutorials, installation hints and these tools' download links.

There is set of tools for each step: i) High-level language compilers and editors (GCC, Emacs); ii) Shell scripting languages and commands (bash, awk, sed, grep etc. ); iii) Performance monitors, system calls, shell scripts and scripting languages (PAPI, Perl, bash, perfmon, getrusage(), gettimeofday() etc.); iv) Plotting tools (GNUPlot); v) Text editing tools (Texlive, Kile, MikTex etc.).

Hint: Everything can be automatic! You will spend a lot of time to learn all those tools, but after that you will do a million experiments with just one code line! I think that if you are reading this document, you're starting a research that will take at least one year of your life.

If you want to change from MS Visual Studio, MS Excel, MS Word, .bat files and discover how to do all that stuff on Linux using GCC, Latex, Bash, GNUplot etc.) I have something to help you start. This is not a Linux course, but it will help you to see an implementation in pure C, make some experiments, learn a bit of bash, plot some graphics and all the things you need to do a research on Linux. This example kit I call it "Things that someone should told me an year ago"!:-). It is composed of a simple C program with 3 bash scripts, that can magically compile, test ... and plot a graphic with just this line of code: ./magic.bash.

Example Kit (without output)

Example Kit (with output)