Introduction Frequency domain operations on spatial or time data are often used as a means for accelerating calculations. For example, when filtering a signal, one can use convolution to perform the operation in the spatial domain or transform the signal…
Blog Archives
Scalable Parallel Multiplication of Big Matrices
Introduction Multiplication of matrices is a fundamental operation in linear algebra and is applied in many engineering and scientific areas. The popular SGEMM function is an example of a standard API used for implementing this operation in its most general…
Hello World! (SPMD version)
Abstract: The purpose of this trivial example is to show that Adapteva’s Epiphany architecture supports a Single Program Multiple Data (SPMD) programming model. Code: #include <stdio.h>; #include <e_coreid.h>; #define MASTER 0×824; int main(void) { //Get the coordinate (core ID) of…
Posted in SPMD
An OpenCL Matrix Multiplication Recipe for Epiphany
Abstract: The examples shows how to use OpenCL to implement an efficient matrix multiplication routine on the Epiphany architecture. To reduce off-chip data movements, our OpenCL SDK was agumented with a few key extension calls to allow for more efficient…
Tagged with: benchmark, matrix multiplication, opencl, optimized, parallel, white paper
Posted in OpenCL
Posted in OpenCL
Matrix Multiplication Optimization
Abstract: The examples shows two ways of performing matrix multiplication, a simple one that gives moderate performance, and a slightly more complex versions that achieves optimal performance using the e-gcc compiler. The optimal code was partially unrolled to allow the…
Posted in ANSI-C
Hello World!
Abstract: The purpose of this trivial example is to show that Adapteva’s Epiphany architecture is a true ANSI-C programmable processor. Whether the processor is being used as an embedded multicore processor or in a high performance PCI express accelerator configuration,…
