RT info:eu-repo/semantics/conferenceObject
T1 One Tier Dataflow Programming Model for Hybrid Distributed- and Shared-Memory Systems
A1 Fresno Bausela, Javier
A1 González Escribano, Arturo
A1 Llanos Ferraris, Diego Rafael
AB Dataflow programming consists in developing a program by describing its sequential stages and the interactions between them. The runtimes supporting this kind of programming are responsible of exploiting the parallelism by concurrently executing the different stages when their dependencies have been met. In this paper we introduce a new parallel programming model and framework based on the dataflow paradigm. Its features are: It is a unique one-tier model that supports hybrid shared- and distributed-memory systems; it can express activities arbitrarily linked, including cycles; it uses a distributed work-stealing mechanism to allow Multiple-Producer/Multiple-Consumer configurations; and it has a run-time mechanism for the reconfiguration of the dependences network which also allows to create task-to-taskaffinities. We present an evaluation using examples of different classes of applications. Experimental results show that programs generated using this framework deliver good performance, and that the new abstractions introduce minimal overheads.
PB Universidad de Valladolid, Escuela de Ingeniería Informática
YR 2016
FD 2016
LK http://uvadoc.uva.es/handle/10324/29119
UL http://uvadoc.uva.es/handle/10324/29119
LA eng
NO HiPEAC 2016 Workshop on High-Level Parallel Programming for GPUs (HLPGPU), Prague, Jan. 18-20 2016.
NO Producción Científica
DS UVaDOC
RD 27-abr-2024