GithubHelp home page GithubHelp logo

iotest's Introduction

iotest

Motivation

  • One tool for different use cases.
  • Simple output format to parse for further analysis.
  • Creating own tests for better understanding how IO works.
  • Other tools have restricted support for using greater random data blocks...
    • IOzone is restricted to 16M record size.
    • Creating random data with DD can take huge amount of time or workarounds need to be done.
  • Direct IO is not always possible e.g. on ZFS, so big record size with random data should be used when compression is activated.
  • Random generated data used for tests should be hardly compressible, it's creations should be fast and the data useable again.
  • IOzone always makes a write+rewrite on a specific filepath (on Lustre a rewrite will create a new file instead keeping the IO on the previous specified stripe settings for first created file).
  • fio creates new file, which will drop previous specified stripping information of a file and most likely result in a file on a different OST.

Build

Make static build so it can be used on other machines without compiling again, but with same architecture:

g++ src/iotest.cpp -o bin/iotest -ansi -pedantic -Wall -Wextra -std=c++17 --static

Usage

Help

./bin/iotest -h

USAGE:
-b BLOCK SIZE 1-999{KM}/1G
-t TOTAL SIZE 1-999{KMGT}
IO MODES:
   * -r SEQUENTIAL READ
   * -w SEQUENTIAL WRITE
   * -R RANDOM READ
   * -W RANDOM WRITE
-f FILEPATH
-s SEED NUMBER
-y SYNC ON WRITE
-v PRINT VERSION

Example for Synchronous Sequential Write

./bin/iotest -w -y -b 1M -t 1G -f /media/hdd/test.tmp

Note:

For random write mode other seed should be used rather than used for sequential write with same block size, otherwise same block data will be written, which might be optimized on writes.

Output Format

Fields:

  • start_timestamp (format: "YYYY-MM-DD HH:MM:SS")
  • end_timestamp (format: "YYYY-MM-DD HH:MM:SS")
  • elapsed_time (seconds)
  • throughput (MB/s)
  • description (depending on the test some meta infromation is provided)

The output fields are pipe separated:

2022-03-04 12:03:56|2022-03-04 12:05:09|73|14|sequential-sync-write-1M-1G|/media/hdd/test.tmp|

iotest's People

Contributors

gabrieleiannetti avatar abhilasha2801 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.