Non Linear Divisible Loads: There is No Free Lunch
BEAUMONT, Olivier
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
LARCHEVÊQUE, Hubert
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
MARCHAL, Loris
Optimisation des ressources : modèles, algorithmes et ordonnancement [ROMA]
Laboratoire de l'Informatique du Parallélisme [LIP]
Optimisation des ressources : modèles, algorithmes et ordonnancement [ROMA]
Laboratoire de l'Informatique du Parallélisme [LIP]
BEAUMONT, Olivier
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
LARCHEVÊQUE, Hubert
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
MARCHAL, Loris
Optimisation des ressources : modèles, algorithmes et ordonnancement [ROMA]
Laboratoire de l'Informatique du Parallélisme [LIP]
< Reduce
Optimisation des ressources : modèles, algorithmes et ordonnancement [ROMA]
Laboratoire de l'Informatique du Parallélisme [LIP]
Language
en
Communication dans un congrès
This item was published in
IPDPS 2013, 27th IEEE International Parallel & Distributed Processing Symposium, 2013-05, Boston. 2013
English Abstract
Divisible Load Theory (DLT) has received a lot of attention in the past decade. A divisible load is a perfect parallel task, that can be split arbitrarily and executed in parallel on a set of possibly heterogeneous resources. ...Read more >
Divisible Load Theory (DLT) has received a lot of attention in the past decade. A divisible load is a perfect parallel task, that can be split arbitrarily and executed in parallel on a set of possibly heterogeneous resources. The success of DLT is strongly related to the existence of many optimal resource allocation and scheduling algorithms, what strongly differs from general scheduling theory. Moreover, recently, close relationships have been underlined between DLT, that provides a fruitful theoretical framework for scheduling jobs on heterogeneous platforms, and MapReduce, that provides a simple and efficient programming framework to deploy applications on large scale distributed platforms. The success of both have suggested to extend their framework to non-linear complexity tasks. In this paper, we show that both DLT and MapReduce are better suited to workloads with linear complexity. In particular, we prove that divisible load theory cannot directly be applied to quadratic workloads, such as it has been proposed recently. We precisely state the limits for classical DLT studies and we review and propose solutions based on a careful preparation of the dataset and clever data parti- tioning algorithms. In particular, through simulations, we show the possible impact of this approach on the volume of communications generated by MapReduce, in the context of Matrix Multiplication and Outer Product algorithms.Read less <
English Keywords
Divisible Load Theory
MapReduce
Scheduling
Resource Allocation
Matrix Multiplica- tion
Sorting
ANR Project
Simulation de systèmes de prochaine génération - ANR-11-INFR-0013
Origin
Hal imported