Optimized Composition: Generating Efficient Code for Heterogeneous Systems from Multi-Variant Components, Skeletons and Containers

TL;DR

This survey reviews recent frameworks for high-level, portable programming of heterogeneous multi-core systems, focusing on components, skeletons, and containers, and discusses optimization challenges for efficient code generation and execution.

Contribution

It provides a comprehensive overview of current approaches and highlights key optimization problems in generating efficient code for heterogeneous systems.

Findings

Identifies main frameworks and their features

Discusses optimization challenges in code generation

Highlights future research directions

Abstract

In this survey paper, we review recent work on frameworks for the high-level, portable programming of heterogeneous multi-/manycore systems (especially, GPU-based systems) using high-level constructs such as annotated user-level software components, skeletons (i.e., predefined generic components) and containers, and discuss the optimization problems that need to be considered in selecting among multiple implementation variants, generating code and providing runtime support for efficient execution on such systems.