Name | Modified | Size | Downloads / Week |
---|---|---|---|
comparativeBenchmarks-1.0.5-src.zip | 2013-08-23 | 905.7 kB | |
README.md | 2013-08-23 | 1.3 kB | |
comparativeBenchmarks-1.0.4-src.zip | 2013-02-21 | 791.2 kB | |
comparativeBenchmarks-1.0.3-src.zip | 2012-12-04 | 789.6 kB | |
comparativeBenchmarks-1.0.2-src.zip | 2012-10-19 | 788.6 kB | |
comparativeBenchmarks-1.0.1-src.zip | 2012-10-05 | 788.1 kB | |
comparativeBenchmarks-1.0.0-src.zip | 2012-08-28 | 787.4 kB | |
Totals: 7 Items | 4.9 MB | 0 |
Project Home
Versions
1.0.5 - Updated some of the Java benchmarks to build properly with the current version of the timing and profiling libraries.
1.0.4 - Updated makefiles to match the current CPoorMansTimingAndProfiling use of the traditional lib/ and include/ layout
1.0.3 - Modified the Parallelism tests to also measure task activation times
1.0.2 - Removed dependence on jcmdline from many of the benchmarks - Fixed minor though annoying bug in Parallelism-Benchmarks-J
1.0.1 - Minor cleanup and tweaking - A Makefile issue cleaned up in some of the benchmarks
1.0.0 - Initial Release
Overview
This is a collection of software benchmarks developed to compare the performance of Java with C on "identical" code. No language libraries were used in the timed parts of the tests to avoid implementation differences. Some of the benchmarks are also implemented in Python and Scala.
Benchmarks include:
- Bit twiddling
- Numerical computing
- In-memory data structure creation and walking
- Compute parallelism
- Calling out from Java/Python/Jython to native libraries
- Utilization of graphics processing units (GPU) from Java and C