A Distributed Shared Memory Model and C++ Templated Meta-Programming Interface for the Epiphany RISC Array Processor

https://doi.org/10.1016/j.procs.2017.05.221Get rights and content
Under a Creative Commons license
open access

Abstract

The Adapteva Epiphany many-core architecture comprises a scalable 2D mesh Network-on-Chip (NoC) of low-power RíSC cores with minimal uncore functionality. Whereas such a processor offers high computational energy efficiency and parallel scalability, developing effective programming models that address the unique architecture features has presented many challenges. We present here a distributed shared memory (DSM) model supported in software transparently using C++ templated meta-programming techniques. The approach offers an extremely simple parallel programming model well suited for the architecture. Initial results are presented that demonstrate the approach and provide insight into the efficiency of the programming model and also the ability of the NoC to support a DSM without explicit control over data movement and localization.

Keywords

2D RISC array
Adapteva Epiphany
Parallella
distributed shared memory
templated meta-programming
parallel programming model

Cited by (0)