- Parallel colt was ommitted due to some difficulties in running it
- Benchmark run by Anders Peterson
Links to the results:
- Summary Results
- Pure Java Results
- Java and Native Results
|Date||2015 / 07|
|OS||Mac OS X 10.10.4 64-bit|
|CPU||2x2.26 Quad-Core Intel Xeon. 16-threads|
|CPU Cache||8 MB L3, 256kB L2, 32kB L1|
|JVM||Oracle Java HotSpot(TM) 64-Bit Server 1.8.0_45|
The following results are a weighted sum across all operations within each matrix size. Operations which take longer will have more weight. If a library could not finish an operation then its score is set to zero.
__NOTE__ The weight is computed from the amount of time the fastest library takes to complete. Which is why the results change a bit from Java to Java + Native.
These results show the performance of libraries that have code written entirely in Java.
These results show the performance of libraries that either use pure Java or calls to native libraries.