Skip to content

lkno0705/MatrixMultiplication

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MatrixMultiplication

Just a little playground, to test and try the benefits of Running Calculations on CPU or GPU with multiple threads.

System stats for the compared run:

Default matrix size: 1440x1440
Thread count: 24
CPU: Ryzen 9 3900X
GPU: Nvidia RTX 2070Super

Java:

Type int

Single: 7.769s
Threaded: 1.124s

Type float

Single: 6.837s
Threaded: 1.225s

C++

Single: 2.21483s
Threaded: 0.460356s

SIMD

Single: 0.5412s
Threaded: 0.1225s

CUDA

Single: -
Threaded: 21.499 ms

Python

Single: 230.283s
Threaded: 17.304s

Kotlin

Single: 6.5s
Threaded: 1.15s

Rust

Single: 2.103s
Threaded: 0.496s

C#

Single: 3.16s
Threaded: 0.515s

C

Single: 2.177s
Threaded: 0.456s

GO

Single: 2.556s
Threaded: 0.4934s

x64 NASM

Single: 0.621s
Threaded: 0.100s

PHP

Single: 160.47s
Threaded: -

Julia

Single: 3.171s
Threaded: 0.576

Excel VBA (Matrix size: 340x340)

Single: 936s
Threaded: - (Not supported)

XSLT (Matrix size: 340x340)

Firefox: 1 801s
Chrome: 4 101s

F#

Single: 2.923s
Threaded: 0.518s

JS

Single: 14.25618s
Threaded: -

OpenCL

Single: To be tested
Threaded: To be tested

SQLite (Matrix size: 340x340)

Single: 41.61s
Threaded: -

TCL

Single: -
Threaded: 170,69s

Lua

Single: To be tested
Threaded: To be tested

Zig

Single: To be tested
Threaded: To be tested