FlameGraph VS pmu-tools

We'll use async-profiler and flame graphs for profiling. To simplify the process, we'll run the code using JBang.

And flame graphs excel and this kind of thing

https://www.brendangregg.com/flamegraphs.html

which can output in a format understood by Brendan Gregg's flame frames (https://www.brendangregg.com/flamegraphs.html)

But that's not quite the kind of tracing you're talking about. We also built a printf-style interface to our recording files, which seems closer:

Strategies included discontinuing or resizing underutilized services, transitioning to more cost-effective solutions, reducing the current resources to the amount of resources that we need for our application, and conducting detailed analyses of computing resource utilization through tools like flamegraphs. This detailed scrutiny helped identify and rectify significant cost-driving areas, such as garbage collection and application configurations.

Flame Graphs by Brendan Gregg

Historically brendangregg's since AIUI he basically invented flamegraphs

https://www.brendangregg.com/flamegraphs.html

So if you can make your tool eat whatever https://github.com/brendangregg/FlameGraph is fed with you're going to support a lot of existing tooling across OSes and languages.

“Flame graphs are a visualization of hierarchical data, created to visualize stack traces of profiled software so that the most frequent code-paths to be identified quickly and accurately.”

SVGs are amazing for interactive visualisation too. Like Flamegraphs: https://www.brendangregg.com/flamegraphs.html

This may be a good example of the application of a flame graph but it is not a good demonstration of flame graphs; the graph is nearly incidental. The source has an actual explanation.

A profiling flame graph from Graphile Crystal (a precursor to Grafast) using GraphQL.js' executor (each tick is 1ms, total: 29ms). As we removed more and more responsibilities from GraphQL.js, we ended up only using it for output. Replacing this final responsibility with a custom implementation in Graphile Crystal itself, we reduced execution time for this query down to 15.5ms (effectively removing the majority of the yellow portion of the flame graph).

I am not seeing it mentioned anywhere, but for people looking for a good starting point on "low-level" CPU performance debugging, intel's CPU top-down u-architecture method (https://www.intel.com/content/www/us/en/docs/vtune-profiler/...) is a good systematic way to understand where you CPU is speeding most of it's cycle.

They also have two tools which basically implement this analysis and spit a bunch of very useful metric that are actionable and very easy to understand

- Intel Vtune is a fantastic tool to start with. It's currently free to use, support most OSes and very friendly to use for beginner.

- Intel pmu-tools (https://github.com/andikleen/pmu-tools) is basically command line version of Vtune.

Install some tool that would help you see the performance of your system, like a graph of the CPU usage, the top processes being used, disk activity/read/write, etc. Every time you run your program, glance at those numbers, eventually you'll develop an intuition. Basically write code and profile. A good exercise would be practicing with data structures, this site has an exhaustive list of them, find some stuff that's interesting then google the implementation, then build it yourself, test it, debug, profile, optimize, and understand the performance constraints. Eventually you'll develop better understanding and can compare between other people's works, optimizing them. If you want to go beyond, read some papers on lock-free algorithms https://github.com/JCTools/JCTools/tree/master/resources then read Brendan Gregg's blog and books. Read about how profiling tools work https://github.com/andikleen/pmu-tools/wiki/toplev-manual

Toplev is a godsend (thank you Andi Kleen!). If you work with perf you'll love this.

https://github.com/andikleen/pmu-tools