DocumentCode :
656225
Title :
The DEEP Project - Pursuing Cluster-Computing in the Many-Core Era
Author :
Eicker, Norbert ; Lippert, Thomas ; Moschny, Thomas ; Suarez, Estela
Author_Institution :
Julich Supercomput. Centre, Forschungszentrum Julich GmbH, Julich, Germany
fYear :
2013
fDate :
1-4 Oct. 2013
Firstpage :
885
Lastpage :
892
Abstract :
Homogeneous cluster architectures dominating high-performance computing (HPC) today are challenged, in particular when thinking about reaching Exascale by the end of the decade, by heterogeneous approaches utilizing accelerator elements. The DEEP (Dynamical Exascale Entry Platform) project aims for implementing a novel architecture for high-performance computing consisting of two components - a standard HPC Cluster and a cluster of many-core processors called Booster. In order to make the adaptation of application codes to this Cluster-Booster architecture as seamless as possible, DEEP provides a complete programming environment. It integrates the offloading functionality given by the MPI standard with an abstraction layer based on the task-based OmpSs programming paradigm. This paper presents the DEEP project with an emphasis on the DEEP programming environment.
Keywords :
application program interfaces; message passing; multiprocessing systems; parallel programming; Booster cluster; DEEP programming environment; DEEP project; MPI standard; abstraction layer; accelerator elements; application codes; cluster-booster architecture; cluster-computing; dynamical exascale entry platform; high-performance computing; homogeneous cluster architectures; many-core era; many-core processors; message passing interface; offloading functionality; programming environment; standard HPC cluster; task-based OmpSs programming paradigm; Computer architecture; Fabrics; Hardware; Parallel processing; Program processors; Programming; Scalability; Computer Architecture; Exascale; HPC;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Parallel Processing (ICPP), 2013 42nd International Conference on
Conference_Location :
Lyon
ISSN :
0190-3918
Type :
conf
DOI :
10.1109/ICPP.2013.105
Filename :
6687429
Link To Document :
بازگشت