Mapping tasks to processors in heterogeneous multiprocessor architectures

Título:

Mapping tasks to processors in heterogeneous multiprocessor architectures:  the MATEHa algorithm

Autor:

De Giusti, Laura Cristina

Colaboradores:

; De Giusti, Armando Eduardo; Naiouf, Ricardo Marcelo; Chichizola, Franco

Temas:

MULTIPROCESADORES

En:

International Conference on the Chilean Computer Science Society (27th : 2008 : Punta Arenas)

Resumen:

An automatic task-to-processor mapping algorithm is analyzed in parallel systems that run over loosely coupled distributed architectures. This research is based on the TTIGHa model that allows predicting parallel application performance running over heterogeneous architectures. In particular, the heterogeneity of both processors communications is taken into consideration. the results obtained with the TTIGHa model, the MATEHa algorithm for task-to-processors assignment is presented its implementation is analyzed. Experimental results working on subsets of twocluster heterogeneous machines are presented, analyzing the resulting mapping scheme with MATEHa two previous mapping methods: MATE HEFT. Finally, the algorithm robustness is considered based on the variation of model parameters: interprocess communication times processing times.

URL/DOI:

Medio:

Soporte electrónico

Tipo de documento:

Artículo

Idioma:

Español

Publicación:

, 2008

Puede solicitar más fácilmente el ejemplar con: A0230

Ver estantes