performance.tools.md

C++ links: performance tools

Benchmarking

• benchmark (Google)
  • https://github.com/google/benchmark
  • https://opensource.googleblog.com/2014/01/introducing-benchmark.html
• Celero
  • https://github.com/DigitalInBlue/Celero
• hayai - the C++ benchmarking framework
  • https://github.com/nickbruun/hayai
• moodycamel::microbench
  • https://github.com/cameron314/microbench
• geiger: A micro benchmark library in C++ that supports hardware performance counters
  • https://github.com/david-grs/geiger
• Nonius
  • https://nonius.io/

Examples:

• http://www.bfilipek.com/2016/01/micro-benchmarking-libraries-for-c.html
• http://www.bfilipek.com/2016/02/revisiting-old-benchmark-vector-of.html
• http://www.bfilipek.com/2016/05/google-benchmark-library.html

Systems Benchmarking Crimes - Gernot Heiser - https://www.cse.unsw.edu.au/~gernot/benchmarking-crimes.html

• Tools for microarchitectural benchmarking - https://dendibakh.github.io/blog/2018/04/03/Tools-for-microarchitectural-benchmarking

Optimization

• MAQAO (Modular Assembly Quality Analyzer and Optimizer)
  • http://www.maqao.org/
  • http://maqao.bordeaux.inria.fr/

Profiling

• Agner Fog's test programs for measuring clock cycles and performance monitoring - http://www.agner.org/optimize/#testp

• BCC - Tools for BPF-based Linux IO analysis, networking, monitoring, and more

  • https://iovisor.github.io/bcc/
  • https://github.com/iovisor/bcc
  • https://github.com/iovisor/bpf-docs
  • http://www.brendangregg.com/blog/2016-03-05/linux-bpf-superpowers.html
  • http://www.brendangregg.com/blog/2016-03-28/linux-bpf-bcc-road-ahead-2016.html
  • http://www.brendangregg.com/blog/2016-06-14/ubuntu-xenial-bcc-bpf.html
  • https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf/

• Event Tracing for Windows (ETW) / Windows Performance Toolkit – Xperf

  • https://blogs.msdn.microsoft.com/ntdebugging/2008/04/03/windows-performance-toolkit-xperf/
  • ETW Central: https://randomascii.wordpress.com/2015/09/24/etw-central/

• gperftools (originally Google Performance Tools)
  "The fastest malloc we’ve seen; works particularly well with threads and STL. Also: thread-friendly heap-checker, heap-profiler, and cpu-profiler."

  • https://github.com/gperftools/gperftools

• gprof2dot
  "Python script to convert the output from many profilers into a dot graph."

  • https://github.com/jrfonseca/gprof2dot

• Intel Architecture Code Analyzer (IACA) - https://software.intel.com/en-us/articles/intel-architecture-code-analyzer

• Intel Performance Counter Monitor (PCM)

• http://www.intel.com/software/pcm
• https://intel-pcm-api-documentation.github.io/

• Likwid: Performance monitoring and benchmarking suite

  • https://github.com/RRZE-HPC/likwid
  • https://github.com/RRZE-HPC/likwid/wiki
  • https://github.com/RRZE-HPC/likwid/wiki/TutorialStart
  • https://github.com/RRZE-HPC/likwid/wiki/likwid-perfctr
  • https://github.com/RRZE-HPC/likwid/wiki/TutorialMarkerC
  • https://github.com/RRZE-HPC/likwid/wiki/PatternsHaswellEP

• perf

  • https://perf.wiki.kernel.org/
  • http://www.brendangregg.com/perf.html
  • https://github.com/torvalds/linux/tree/master/tools/perf/Documentation
  • https://suchakra.wordpress.com/2016/06/17/deconstructing-perfs-data-file/
  • http://www.brendangregg.com/linuxperf.html

• perf-tools

  • https://github.com/brendangregg/perf-tools

• perf_events: The Unofficial Linux Perf Events Web-Page - http://web.eece.maine.edu/~vweaver/projects/perf_events/

• perfmon2 - http://perfmon2.sourceforge.net/

  • "Perfmon2 aims to be a portable interface across all modern processors. It is designed to give full access to a given PMU and all the corresponding hardware performance counters. Typically the PMU hardware implementations use a different number of registers, counters with different length and possibly other unique features, a complexity that the software has to cope with. Although processors have different PMU implementations, they usually use configurations registers and data registers. Perfmon2 provides a uniform abstract model of these registers and exports read/write operations accordingly."

• Performance Application Programming Interface (PAPI)

  • http://icl.cs.utk.edu/papi/
  • http://icl.cs.utk.edu/projects/papi/wiki/Main_Page
  • http://www.drdobbs.com/tools/performance-monitoring-with-papi/184406109
  • papi-wrapper (C++ library) - https://github.com/sean-chester/papi-wrapper

• pmu tools: Intel PMU profiling tools

  • https://github.com/andikleen/pmu-tools
  • https://github.com/andikleen/pmu-tools/wiki/toplev-manual
  • pmu-tools part I - introduction, ocperf - http://halobates.de/blog/p/245
  • pmu-tools part II - toplev - http://halobates.de/blog/p/262

• sysdig

  • https://github.com/draios/sysdig
  • https://sysdig.com/blog/50-shades-of-system-calls/

Profiling - Memory

• Heaptrack - A Heap Memory Profiler for Linux

  • https://github.com/KDE/heaptrack
  • http://milianw.de/blog/heaptrack-a-heap-memory-profiler-for-linux

• memusage - profile memory usage of a program

  • http://man7.org/linux/man-pages/man1/memusage.1.html

• Typegrind
  a type preserving heap profiler for C++ - collects memory allocation information with type information

  • https://typegrind.github.io/
  • https://github.com/typegrind/typegrind

Visualization

• pprof - a tool for visualization and analysis of profiling data
  • https://github.com/google/pprof
• Flame Graphs
  • http://www.brendangregg.com/flamegraphs.html
  • http://queue.acm.org/detail.cfm?id=2927301
• FlameScope: a visualization tool for exploring different time ranges as Flame Graphs
  • https://github.com/Netflix/flamescope
  • Netflix FlameScope - https://medium.com/@NetflixTechBlog/netflix-flamescope-a57ca19d47bb