↓
 
  • Phoronix
  • gromacs
  • OpenFOAM

Performance analysis, tools and experiments

An eclectic collection

Performance analysis, tools and experiments
  • Home
  • Blog
  • Tools
    • wspy – workload spy
  • Workloads
    • Geekbench
    • gromacs
      • lysozyme tutorial
      • PRACE benchmark
    • OpenFOAM
    • Phoronix
      • aobench
      • apache
      • asmfish
      • blake2
      • blender
      • botan
      • build-gcc
      • build-linux-kernel
      • build-llvm
      • build-php
      • bullet
      • c-ray
      • cachebench
      • compilebench
      • compress-p7zip
      • compress-pbzip2
      • compress-zstd
      • ebizzy
      • encode-flac
      • encode-mp3
      • ffmpeg
      • ffte
      • fftw
      • fhourstones
      • fio
      • fs-mark
      • gimp
      • git
      • go-benchmark
      • graphics-magick
      • hackbench
      • himeno
      • hmmer
      • indigobench
      • java-gradle-perf
      • java-scimark2
      • luajit
      • m-queens
      • mafft
      • n-queens
      • nginx
      • numpy
      • octave-benchmark
      • openssl
      • osbench
        • osbench – create processes
        • osbench – create threads
        • osbench – memory
      • parboil
      • pgbench
      • phpbench
      • polybench-c
      • povray
      • primesieve
      • pybench
      • radiance
      • rbenchmark
      • redis
      • rodinia
      • scikit-learn
      • scimark2
      • sqlite
      • stockfish
      • stream
      • stress-ng
      • tensorflow
      • tinymembench
      • tjbench
      • tscp
      • ttsiod-renderer
      • vpxenc
      • x264
      • y-cruncher
  • Experiments
Home→Tags top-down

Tag Archives: top-down

topdown equivalent metrics for AMD Zen?

Performance analysis, tools and experiments Posted on 2018-06-08 by mev2018-06-11

Intel processors provide a useful set of performance counters for doing topdown analysis, particularly at the first topdown level. This first level asks questions in the following hierarchy: Is a Uop dispatched? If yes, is it retired? If yes –> … Continue reading →

Posted in analysis | Tagged analysis technique, performance counters, top-down | Leave a reply

100+ phoronix benchmarks listed

Performance analysis, tools and experiments Posted on 2018-06-03 by mev2018-06-03

On the upper left of the web site is now an item labeled “phoronix”. Clicking there returns a table with a long list of Phoronix benchmarks. For each I’ve recorded the description, scores when running on my Intel reference system, … Continue reading →

Posted in analysis, featured | Tagged phoronix, top-down | Leave a reply

topdown tool – adding support for level 3 backend, rusage

Performance analysis, tools and experiments Posted on 2018-05-25 by mev2018-05-25

I have updated the topdown wrapper script to add some additional cache statistics for level 3 backend related information. I have also added a -x option to display some rusage information.

Continue reading →
Posted in tools | Tagged metrics, top-down | Leave a reply

topdown tool – adding support for level 3 frontend

Performance analysis, tools and experiments Posted on 2018-05-24 by mev2018-05-24

The Intel topdown performance analysis method separates out frontend stalls (level 1) into latency (level 2) and bandwidth (level 2). A frontend latency stall is when 0 uops are issued at all because the frontend is waiting. A frontend bandwidth … Continue reading →

Posted in tools | Tagged metrics, top-down | Leave a reply

topdown tool

Performance analysis, tools and experiments Posted on 2018-05-21 by mev2018-05-21

I have added a basic tool “topdown” that is a wrapper to collect topdown metrics.

Continue reading →
Posted in tools | Tagged metrics, top-down | Leave a reply

wspy – memory analysis for processes; need to sanity check the metrics…

Performance analysis, tools and experiments Posted on 2018-04-29 by mev2018-04-30

As a followup to this post, I’ve implemented per-process capture of backend counters. I can now create a memory report from a process tree.

Continue reading →
Posted in analysis, experiments, tools | Tagged memory, metrics, top-down | Leave a reply

wspy – added topdown memory analysis

Performance analysis, tools and experiments Posted on 2018-04-25 by mev2018-04-25

As a followup to a previous post, I’ve added support to wspy for topdown analysis for backend stalls.

Continue reading →
Posted in tools | Tagged memory, top-down, wspy | 1 Reply

Top down performance counter analysis (part 3) – wspy

Performance analysis, tools and experiments Posted on 2018-04-17 by mev2018-04-19

As described in top down performance counter analysis part 1, top down analysis is an approach that uses key performance counters to characterize an application and then successively drills down with further refinement. On Intel x86 processors, this first level … Continue reading →

Posted in tools | Tagged performance counters, top-down, wspy | Leave a reply

Top down performance counter analysis (part 2) – wspy

Performance analysis, tools and experiments Posted on 2018-04-16 by mev2018-04-17

As described in top down performance counter analysis part 1, top down analysis is an approach that uses key performance counters to characterize an application and then successively drills down with further refinement. On Intel x86 processors, this first level … Continue reading →

Posted in tools | Tagged analysis technique, performance counters, top-down | Leave a reply

Top down performance counter analysis (part 1) – likwid & perf

Performance analysis, tools and experiments Posted on 2018-04-14 by mev2018-04-16

In this posting I summarize top-down performance counter analysis to evaluate workloads and show how this can be measured on Haswell using likwid-perfctr and perf. In part 2 to follow, I’ll describe how top down metrics have been added to … Continue reading →

Posted in experiments | Tagged analysis technique, likwid-perfctr, perf, performance counters, top-down | 1 Reply

Archives

  • March 2020
  • August 2018
  • June 2018
  • May 2018
  • April 2018
  • March 2018

Tags

/proc/stat analysis technique c-ray ftrace geekbench geekbench-description gromacs hyperthreading ipc kernelshark likwid likwid-perfctr likwid-topology memory metrics on_core on_cpu OpenFOAM parboil perf performance counters periodic sampling phoronix phoronix benchmark article phoronix cpu suite pmu_tools process tree progress report ptrace rusage Ryzen spectre system hang top-down trace-cmd virtualization wspy

Meta

  • Log in
  • Entries feed
  • Comments feed
  • WordPress.org
©2023 - Performance analysis, tools and experiments - Weaver Xtreme Theme
↑