DocumentCode :
3199105
Title :
Cashmere: Heterogeneous Many-Core Computing
Author :
Hijma, Pieter ; Jacobs, Ceriel J. H. ; Van Nieuwpoort, Rob V. ; Bal, Henri E.
Author_Institution :
VU Univ., Amsterdam, Netherlands
fYear :
2015
fDate :
25-29 May 2015
Firstpage :
135
Lastpage :
145
Abstract :
New generations of many-core hardware become available frequently and are typically attractive extensions for data-centers because of power-consumption and performance benefits. As a result, supercomputers and clusters are becoming heterogeneous and start to contain a variety of many-core devices. Obtaining performance from a homogeneous cluster-computer is already challenging, but achieving it from a heterogeneous cluster is even more demanding. Related work primarily focuses on homogeneous many-core clusters. In this paper we present Cashmere, a programming system for heterogeneous many-core clusters. Cashmere is a tight integration of two existing systems: Satin is a programming system that provides a divide- and-conquer programming model with automatic load-balancing and latency-hiding, while Many-Core Levels is a programming system that provides a powerful methodology to optimize computational kernels for varying types of many-core hardware. We evaluate our system with several classes of applications and show that Cashmere achieves high performance and good scalability. The efficiency of heterogeneous executions is comparable to the homogeneous runs and is >90% in three out of four applications.
Keywords :
divide and conquer methods; multiprocessing systems; programming; resource allocation; workstation clusters; Cashmere; automatic load-balancing; data-centers; divide-and-conquer programming model; heterogeneous cluster; heterogeneous many-core clusters; heterogeneous many-core computing; homogeneous cluster-computer; latency-hiding; many-core devices; many-core hardware; many-core levels; power-consumption; programming system; supercomputers; Computational modeling; Graphics processing units; Hardware; Kernel; Parallel processing; Performance evaluation; Programming; cluster; divide-and-conquer; heterogeneous; many-core;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel and Distributed Processing Symposium (IPDPS), 2015 IEEE International
Conference_Location :
Hyderabad
ISSN :
1530-2075
Type :
conf
DOI :
10.1109/IPDPS.2015.38
Filename :
7161503
Link To Document :
بازگشت