2016-05-26 19:30:19 +02:00
# BLAS libraries benchmarks
Andrzej Wójtowicz
2016-11-29 11:48:07 +01:00
Document generation date: 2016-11-29 11:40:07
2016-07-14 17:36:59 +02:00
2016-11-29 11:48:07 +01:00
[![DOI ](https://zenodo.org/badge/doi/10.5281/zenodo.57910.svg )](http://dx.doi.org/10.5281/zenodo.57910)
2016-06-15 16:03:04 +02:00
2016-07-14 17:26:58 +02:00
This document presents timing results for BLAS ([Basic Linear Algebra Subprograms](https://en.wikipedia.org/wiki/Basic_Linear_Algebra_Subprograms)) libraries in [R ](https://en.wikipedia.org/wiki/R_(programming_language )) on diverse CPUs and GPUs.
### Changelog
2016-11-29 11:48:07 +01:00
* 2016-11-29: **results:** added Intel Xeon E3-1275 v5.
2016-11-25 13:06:20 +01:00
* 2016-11-25: **results:** added Intel Atom C2758.
2016-07-14 17:26:58 +02:00
* 2016-07-14: **results:** added Intel Core i5-6500; changed results view of gcbd benchmark to relative performance gain; changed reference CPU (Intel Pentium Dual-Core E5300) and GPU (NVIDIA GeForce GT 630M); **code:** fixed target architecture detection for Intel Core i5-6500-like CPUs in multi-threaded Atlas library; added info how to force target architecture in GotoBLAS2 and BLIS libraries.
2016-05-26 19:30:19 +02:00
## Table of Contents
1. [Configuration ](#configuration )
2. [Results per host ](#results-per-host )
2016-11-29 11:48:07 +01:00
* [Intel Xeon E3-1275 v5 ](#intel-xeon-e3-1275-v5 )
2016-06-06 15:10:29 +02:00
* [Intel Core i7-4790K + MSI GeForce GTX 980 Ti Lightning ](#intel-core-i7-4790k--msi-geforce-gtx-980-ti-lightning )
2016-05-26 19:39:26 +02:00
* [Intel Core i5-4590 + NVIDIA GeForce GT 430 ](#intel-core-i5-4590--nvidia-geforce-gt-430 )
2016-05-31 00:00:52 +02:00
* [Intel Core i5-4590 + NVIDIA GeForce GTX 750 Ti ](#intel-core-i5-4590--nvidia-geforce-gtx-750-ti )
2016-07-14 17:26:58 +02:00
* [Intel Core i5-6500 ](#intel-core-i5-6500 )
2016-05-26 19:30:19 +02:00
* [Intel Core i5-3570 ](#intel-core-i5-3570 )
* [Intel Core i3-2120 ](#intel-core-i3-2120 )
* [Intel Core i3-3120M ](#intel-core-i3-3120m )
2016-05-31 00:00:52 +02:00
* [Intel Core i5-3317U + NVIDIA GeForce GT 630M ](#intel-core-i5-3317u--nvidia-geforce-gt-630m )
2016-11-25 13:06:20 +01:00
* [Intel Atom C2758 ](#intel-atom-c2758 )
2016-06-03 15:40:33 +02:00
* [Intel Pentium Dual-Core E5300 ](#intel-pentium-dual-core-e5300 )
2016-05-26 19:30:19 +02:00
3. [Results per library ](#results-per-library )
* [Netlib ](#netlib )
* [Atlas (st) ](#atlas-st )
* [OpenBLAS ](#openblas )
* [Atlas (mt) ](#atlas-mt )
* [GotoBLAS2 ](#gotoblas2 )
* [MKL ](#mkl )
* [BLIS ](#blis )
2016-05-31 00:00:52 +02:00
* [cuBLAS ](#cublas )
2016-05-26 19:30:19 +02:00
***
## Configuration
2016-07-14 17:26:58 +02:00
**OS**: [Debian ](https://www.debian.org/ ) Jessie, kernel 4.4
2016-06-06 22:47:50 +02:00
2016-06-06 15:07:17 +02:00
**R software**: [Microsoft R Open ](https://mran.microsoft.com/open/ ) (3.2.4)
2016-05-26 19:30:19 +02:00
**Libraries**:
|CPU (single-threaded)|CPU (multi-threaded)|GPU|
|---|---|---|
2016-05-31 00:00:52 +02:00
|[Netlib](http://www.netlib.org/) (debian package, blas 1.2.20110419, lapack 3.5.0)|[OpenBLAS](http://www.openblas.net/) (debian package, 0.2.12)|[NVIDIA cuBLAS](https://developer.nvidia.com/cublas) (NVBLAS 6.5 + Intel MKL)|
|[ATLAS](http://math-atlas.sourceforge.net/) (debian package, 3.10.2)|[ATLAS](http://math-atlas.sourceforge.net/) (dev branch, 3.11.38)| |
| |[GotoBLAS2](https://prs.ism.ac.jp/~nakama/SurviveGotoBLAS2/) (Survive fork, 3.141)| |
| |[Intel MKL](https://mran.microsoft.com/download/) (part of RevoMath package, 3.2.4)| |
| |[BLIS](https://github.com/flame/blis) (dev branch, 0.2.0+/17.05.2016)| |
2016-05-26 19:30:19 +02:00
**Hosts**:
|No.|CPU|GPU|
|---|---|---|
2016-11-29 11:48:07 +01:00
|1.|[Intel Xeon E3-1275 v5](http://ark.intel.com/products/88177/Intel-Xeon-Processor-E3-1275-v5-8M-Cache-3_60-GHz)| - |
|2.|[Intel Core i7-4790K](http://ark.intel.com/products/80807/Intel-Core-i7-4790K-Processor-8M-Cache-up-to-4_40-GHz) (OC 4.5 GHz)|[MSI GeForce GTX 980 Ti Lightning](https://us.msi.com/Graphics-card/GTX-980-Ti-LIGHTNING.html#hero-specification)|
|3.|[Intel Core i5-4590](http://ark.intel.com/products/80815/Intel-Core-i5-4590-Processor-6M-Cache-up-to-3_70-GHz)|[NVIDIA GeForce GT 430](http://www.geforce.com/hardware/desktop-gpus/geforce-gt-430/specifications)|
|4.|[Intel Core i5-4590](http://ark.intel.com/products/80815/Intel-Core-i5-4590-Processor-6M-Cache-up-to-3_70-GHz)|[NVIDIA GeForce GTX 750 Ti](http://www.geforce.com/hardware/desktop-gpus/geforce-gtx-750-ti/specifications)|
|5.|[Intel Core i5-6500](http://ark.intel.com/products/88184/Intel-Core-i5-6500-Processor-6M-Cache-up-to-3_60-GHz)| - |
|6.|[Intel Core i5-3570](http://ark.intel.com/products/65702/Intel-Core-i5-3570-Processor-6M-Cache-up-to-3_80-GHz)| - |
|7.|[Intel Core i3-2120](http://ark.intel.com/products/53426/Intel-Core-i3-2120-Processor-3M-Cache-3_30-GHz)| - |
|8.|[Intel Core i3-3120M](http://ark.intel.com/products/71465/Intel-Core-i3-3120M-Processor-3M-Cache-2_50-GHz)| - |
|9.|[Intel Core i5-3317U](http://ark.intel.com/products/65707/Intel-Core-i5-3317U-Processor-3M-Cache-up-to-2_60-GHz)|[NVIDIA GeForce GT 630M](http://www.geforce.com/hardware/notebook-gpus/geforce-gt-630m/specifications)|
|10.|[Intel Atom C2758](http://ark.intel.com/products/77988/Intel-Atom-Processor-C2758-4M-Cache-2_40-GHz)| - |
|11.|[Intel Pentium Dual-Core E5300](http://ark.intel.com/products/35300/Intel-Pentium-Processor-E5300-2M-Cache-2_60-GHz-800-MHz-FSB)| - |
2016-05-26 19:30:19 +02:00
2016-06-06 15:07:17 +02:00
**Benchmarks**: [R-benchmark-25 ](http://r.research.att.com/benchmarks/R-benchmark-25.R ), [Revolution ](https://gist.github.com/andrie/24c9672f1ea39af89c66#file-rro-mkl-benchmark-r ), [Gcbd ](https://cran.r-project.org/web/packages/gcbd/vignettes/gcbd.pdf ).
2016-05-26 19:30:19 +02:00
# Results per host
2016-11-29 11:48:07 +01:00
## Intel Xeon E3-1275 v5
2016-05-26 19:30:19 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h1_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h1_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h1_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h1_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h1_b3_t4.png)
2016-05-26 19:30:19 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i7-4790K + MSI GeForce GTX 980 Ti Lightning
2016-05-26 19:30:19 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h2_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h2_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h2_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h2_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h2_b3_t4.png)
2016-05-26 19:30:19 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i5-4590 + NVIDIA GeForce GT 430
2016-05-26 19:30:19 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h3_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h3_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h3_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h3_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h3_b3_t4.png)
2016-05-26 19:30:19 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i5-4590 + NVIDIA GeForce GTX 750 Ti
2016-05-26 19:30:19 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_ph_h4_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h4_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h4_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h4_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h4_b3_t4.png)
2016-05-26 19:30:19 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i5-6500
2016-05-31 00:00:52 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-31 00:00:52 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h5_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h5_b3_t1.png)
2016-05-31 00:00:52 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h5_b3_t2.png)
2016-05-31 00:00:52 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h5_b3_t3.png)
2016-05-31 00:00:52 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h5_b3_t4.png)
2016-05-31 00:00:52 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i5-3570
2016-05-31 00:00:52 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-31 00:00:52 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-31 00:00:52 +02:00
![](gen/img/img_ph_h6_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h6_b3_t1.png)
2016-05-31 00:00:52 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h6_b3_t2.png)
2016-05-31 00:00:52 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h6_b3_t3.png)
2016-05-31 00:00:52 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-05-31 00:00:52 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h6_b3_t4.png)
2016-05-31 00:00:52 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i3-2120
2016-06-03 15:40:33 +02:00
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-06-03 15:40:33 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-03 15:40:33 +02:00
![](gen/img/img_ph_h7_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h7_b3_t1.png)
2016-06-03 15:40:33 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h7_b3_t2.png)
2016-06-03 15:40:33 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h7_b3_t3.png)
2016-06-03 15:40:33 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h7_b3_t4.png)
2016-06-03 15:40:33 +02:00
2016-11-29 11:48:07 +01:00
## Intel Core i3-3120M
2016-06-06 15:07:17 +02:00
### R-benchmark-25
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t1.png)
#### Linear regr. over a 2000x2000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t2.png)
#### Eigenvalues of a 600x600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-06-06 15:07:17 +02:00
![](gen/img/img_ph_h8_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h8_b3_t1.png)
#### QR Decomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h8_b3_t2.png)
#### Singular Value Deomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h8_b3_t3.png)
#### Triangular Decomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h8_b3_t4.png)
2016-11-29 11:48:07 +01:00
## Intel Core i5-3317U + NVIDIA GeForce GT 630M
2016-07-14 17:26:58 +02:00
### R-benchmark-25
#### 2800x2800 cross-product matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t1.png)
#### Linear regr. over a 2000x2000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t2.png)
#### Eigenvalues of a 600x600 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b2_t1.png)
#### Cholesky Factorization
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b2_t2.png)
#### Singular Value Deomposition
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b2_t3.png)
#### Principal Components Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b2_t4.png)
#### Linear Discriminant Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h9_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-06 15:07:17 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h9_b3_t1.png)
2016-06-06 15:07:17 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-06 15:07:17 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h9_b3_t2.png)
2016-06-06 15:07:17 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-06 15:07:17 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h9_b3_t3.png)
2016-06-06 15:07:17 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
2016-06-06 15:07:17 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_ph_h9_b3_t4.png)
2016-06-06 15:07:17 +02:00
2016-11-29 11:48:07 +01:00
## Intel Atom C2758
2016-11-25 13:06:20 +01:00
### R-benchmark-25
#### 2800x2800 cross-product matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t1.png)
#### Linear regr. over a 2000x2000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t2.png)
#### Eigenvalues of a 600x600 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-11-29 11:48:07 +01:00
ATLAS (mt) crashes in this test
2016-11-25 13:06:20 +01:00
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-11-29 11:48:07 +01:00
ATLAS (mt) crashes in this test
2016-11-25 13:06:20 +01:00
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b2_t1.png)
#### Cholesky Factorization
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b2_t2.png)
#### Singular Value Deomposition
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b2_t3.png)
#### Principal Components Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b2_t4.png)
#### Linear Discriminant Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h10_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h10_b3_t1.png)
#### QR Decomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h10_b3_t2.png)
#### Singular Value Deomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h10_b3_t3.png)
#### Triangular Decomposition
2016-11-29 11:48:07 +01:00
ATLAS (mt) crashes in this test
2016-11-25 13:06:20 +01:00
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h10_b3_t4.png)
2016-11-29 11:48:07 +01:00
## Intel Pentium Dual-Core E5300
### R-benchmark-25
#### 2800x2800 cross-product matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t1.png)
#### Linear regr. over a 2000x2000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t2.png)
#### Eigenvalues of a 600x600 random matrix
BLIS hangs in this test
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b2_t1.png)
#### Cholesky Factorization
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b2_t2.png)
#### Singular Value Deomposition
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b2_t3.png)
#### Principal Components Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b2_t4.png)
#### Linear Discriminant Analysis
Time in seconds - 10 runs - lower is better
![](gen/img/img_ph_h11_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h11_b3_t1.png)
#### QR Decomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h11_b3_t2.png)
#### Singular Value Deomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h11_b3_t3.png)
#### Triangular Decomposition
Performance gain regarding matrix size - reference: Netlib - from 50 to 5 runs - higher is better
![](gen/img/img_ph_h11_b3_t4.png)
2016-05-26 19:30:19 +02:00
# Results per library
## Netlib
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l1_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l1_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l1_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l1_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l1_b3_t4.png)
2016-05-26 19:30:19 +02:00
## ATLAS (st)
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l2_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l2_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l2_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l2_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l2_b3_t4.png)
2016-05-26 19:30:19 +02:00
## OpenBLAS
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l3_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l3_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l3_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l3_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l3_b3_t4.png)
2016-05-26 19:30:19 +02:00
## ATLAS (mt)
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-11-25 13:06:20 +01:00
Library crashes on Intel Atom C2758 in this test
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-11-25 13:06:20 +01:00
Library crashes on Intel Atom C2758 in this test
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l4_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l4_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l4_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l4_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-11-25 13:06:20 +01:00
Library crashes on Intel Atom C2758 in this test
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l4_b3_t4.png)
2016-05-26 19:30:19 +02:00
## GotoBLAS2
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l5_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l5_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l5_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l5_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l5_b3_t4.png)
2016-05-26 19:30:19 +02:00
## MKL
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l6_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l6_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l6_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l6_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l6_b3_t4.png)
2016-05-26 19:30:19 +02:00
## BLIS
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-11-25 13:06:20 +01:00
Library hangs on Intel Pentium Dual-Core E5300 in this test
2016-06-03 15:40:33 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l7_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l7_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l7_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l7_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: Intel Pentium Dual-Core E5300 - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l7_b3_t4.png)
2016-05-26 19:30:19 +02:00
## cuBLAS
2016-06-06 15:07:17 +02:00
### R-benchmark-25
2016-05-26 19:30:19 +02:00
#### 2800x2800 cross-product matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t1.png)
2016-06-06 15:07:17 +02:00
#### Linear regr. over a 2000x2000 matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t2.png)
2016-06-06 15:07:17 +02:00
#### Eigenvalues of a 600x600 random matrix
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t3.png)
#### Determinant of a 2500x2500 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t4.png)
#### Cholesky decomposition of a 3000x3000 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t5.png)
#### Inverse of a 1600x1600 random matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t6.png)
#### Escoufier's method on a 45x45 matrix
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b1_t7.png)
### Revolution benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b2_t1.png)
#### Cholesky Factorization
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b2_t2.png)
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b2_t3.png)
#### Principal Components Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b2_t4.png)
#### Linear Discriminant Analysis
2016-07-14 17:26:58 +02:00
Time in seconds - 10 runs - lower is better
2016-05-26 19:30:19 +02:00
![](gen/img/img_pl_l8_b2_t5.png)
### Gcbd benchmark
#### Matrix Multiply
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: NVIDIA GeForce GT 630M - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l8_b3_t1.png)
2016-05-26 19:30:19 +02:00
#### QR Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: NVIDIA GeForce GT 630M - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l8_b3_t2.png)
2016-05-26 19:30:19 +02:00
#### Singular Value Deomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: NVIDIA GeForce GT 630M - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l8_b3_t3.png)
2016-05-26 19:30:19 +02:00
#### Triangular Decomposition
2016-07-14 17:26:58 +02:00
Performance gain regarding matrix size - reference: NVIDIA GeForce GT 630M - from 50 to 5 runs - higher is better
2016-05-26 19:30:19 +02:00
2016-07-14 17:26:58 +02:00
![](gen/img/img_pl_l8_b3_t4.png)
2016-04-28 13:38:54 +02:00