https://github.com/jmcph4/jcrl_benchmarks

Performance benchmarks for JCRL
https://github.com/jmcph4/jcrl_benchmarks

Last synced: over 1 year ago
JSON representation

Performance benchmarks for JCRL

• Host: GitHub
• URL: https://github.com/jmcph4/jcrl_benchmarks
• Owner: jmcph4
• License: mit
• Created: 2019-07-13T06:57:10.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2019-07-14T06:08:08.000Z (about 7 years ago)
• Last Synced: 2025-02-16T00:44:16.860Z (over 1 year ago)
• Language: C
• Size: 43 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# JCRL Benchmarks #

---

This repository contains various performance benchmarks for [JCRL](https://github.com/jmcph4/jcrl).

## Building ##

THe benchmarks have two dependencies: JCRL itself (obviously) and [MPH](https://github.com/jmcph4/mph) (in order to run the benchmark script).

    $ git clone git@github.com:jmcph4/jcrl.git

    $ git clone git@github.com:jmcph4/mph.git

    $ git clone git@github.com:jmcph4/jcrl_benchmarks.git

    $ cp jcrl/src/*.h jcrl_benchmarks/include/jcrl -r

    $ cp mph/mph jcrl_benchmarks -r

Building the benchmark executables is straightforward:

    $ make all

This will create executables under `bin/`.

## Running ##

Each benchmarks is runnable as a standalone program; however, the main script is recommended as it handles statistical calculations and automatically runs all of the benchmarks.

    $ python run.py [num_elems] [num_trials]

This will output the benchmark name, mean running time, and the standard deviation of the running time.

Each executable itself outputs two numbers to `stdout`: the number of seconds elapsed and the number of nanoseconds (note this wraps).