This page links to various Linux performance material I've created, including the tools maps on the right, which show: Linux observability tools, Linux benchmarking tools, Linux tuning tools, and Linux observability sar. For more diagrams, see my slide decks below.
- perf_events: (aka "perf") one-liners, examples, visualizations.
- perf-tools: perf analysis tools based on perf_events and ftrace (github).
- ktap: one-liners, examples, and scripts.
- Flame Graphs: using perf, SystemTap, and ktap.
- My lwn.net article Ftrace: The Hidden Light Switch (Aug, 2014).
- Posts about ftrace-based perf-tools: iosnoop for Linux, iosnoop Latency Heat Maps, opensnoop for Linux, execsnoop for Linux.
- Posts about perf_events-based perf-tools: perf Hacktogram
- Posts about perf_events: perf CPU Sampling, perf Static Tracepoints, perf Heat Maps, perf Counting).
- The blog post strace Wow Much Syscall discusses strace(1) for production use, and compares it to advanced tracing tools.
- USE Method: Linux Performance Checklist; also see the USE Method page for the description of this methodology.
- Systems Performance: Enterprise and the Cloud (Prentice Hall, 2013) uses Linux distributions as the primary example.
Previous talks on Linux performance.
LinuxCon North America 2014
At LinuxCon North America I gave a talk on Linux performance tools, summarizing the tool landscape in 50 minutes. My aim was to provide exposure to what exists: it isn't necessary to remember how (you can look that up later), but it is necessary to know that you can. This talk was not videoed.
This was similar to my SCaLE11x talk, but this time I included benchmarking and tuning tools for a more complete summary. I also updated my summary of the tracing tools, and my recent work with ftrace. I wrote a bit more about the talk in a blog post, which links to the blank diagram for you to fill out.
At SCaLE 12x (Southern California Linux Expo, 2014) I gave the keynote on What Linux can learn from Solaris performance and vice-versa. This drew on my experience analyzing Linux vs SmartOS performance, and my work for the Systems Performance book. SmartOS is an OS based on the illumos kernel, which is the active fork of the OpenSolaris kernel.
I've never seen a good talk comparing performance features of both, I suspect in part because it's hard to know them both in enough depth, and also hard to choose from the many differences which should be highlighted.
This presentation also contains ponies. Lots of ponies. These are the unofficial mascots for DTrace, perf_events, SystemTap, ktap, and LTTng, and were designed by the same person (Deirdré) who designed the original (and popular) DTrace ponycorn.
At SCaLE 11x (2013) I gave a talk on Linux performance analysis and tools, and described over twenty observability tools, and then methodologies for using them. During this talk I introduced the Linux performance tools diagram, which I've put at the top of this page (an updated version).
I've used pretty much everything for solving performance issues, including advanced tools like perf, DTrace, and SystemTap, and I was able to explain their role and how they fit together. It was pretty dense: you can treat this as a 60 minute crash course into Linux performance analysis and tools.
Thanks Deirdré Straughan for filming it and then spreading the word afterwards.