A Survey of CPU-GPU Heterogeneous Computing Techniques

[sparsh]

https://www.academia.edu/12355899/A_Sur ... Techniques

Publication: Accepted in ACM Computing Surveys 2015
Author: Sparsh Mittal and Jeffrey Vetter

This paper surveys CPU-GPU heterogeneous systems, at all abstraction layers of system stack, ranging from microarchitecture to system and application-level. It identifies trends in CPU and GPU design (e.g. transistor count, core-count, 3D, interconnect) and also compares fused CPU-GPU chips (e.g. Llano) with discrete GPU systems. It classifies the research works based on the programming language used (e.g. OpenCL, CUDA, OpenMP etc.) and their area of application (e.g. Physics, numerical algebra etc.). The paper also shows benchmarks used for evaluating CPU-GPU systems.

Keywords: CPU-GPU heterogeneous/hybrid/collaborative computing, workload division/partitioning, dynamic/static load-balancing, pipelining, programming frameworks, high performance computing