Non Linear Divisible Loads: There is No Free Lunch
hal.structure.identifier | Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE] | |
hal.structure.identifier | Laboratoire Bordelais de Recherche en Informatique [LaBRI] | |
dc.contributor.author | BEAUMONT, Olivier | |
hal.structure.identifier | Algorithmics for computationally intensive applications over wide scale distributed platforms [CEPAGE] | |
hal.structure.identifier | Laboratoire Bordelais de Recherche en Informatique [LaBRI] | |
dc.contributor.author | LARCHEVÊQUE, Hubert | |
hal.structure.identifier | Optimisation des ressources : modèles, algorithmes et ordonnancement [ROMA] | |
hal.structure.identifier | Laboratoire de l'Informatique du Parallélisme [LIP] | |
dc.contributor.author | MARCHAL, Loris | |
dc.date.accessioned | 2024-04-15T09:44:17Z | |
dc.date.available | 2024-04-15T09:44:17Z | |
dc.date.created | 2012 | |
dc.date.issued | 2013 | |
dc.date.conference | 2013-05 | |
dc.identifier.uri | https://oskar-bordeaux.fr/handle/20.500.12278/197830 | |
dc.description.abstractEn | 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. | |
dc.description.sponsorship | Simulation de systèmes de prochaine génération - ANR-11-INFR-0013 | |
dc.language.iso | en | |
dc.subject.en | Divisible Load Theory | |
dc.subject.en | MapReduce | |
dc.subject.en | Scheduling | |
dc.subject.en | Resource Allocation | |
dc.subject.en | Matrix Multiplica- tion | |
dc.subject.en | Sorting | |
dc.title.en | Non Linear Divisible Loads: There is No Free Lunch | |
dc.type | Communication dans un congrès | |
dc.subject.hal | Informatique [cs]/Calcul parallèle, distribué et partagé [cs.DC] | |
bordeaux.hal.laboratories | Laboratoire Bordelais de Recherche en Informatique (LaBRI) - UMR 5800 | * |
bordeaux.institution | Université de Bordeaux | |
bordeaux.institution | Bordeaux INP | |
bordeaux.institution | CNRS | |
bordeaux.conference.title | IPDPS 2013, 27th IEEE International Parallel & Distributed Processing Symposium | |
bordeaux.country | US | |
bordeaux.conference.city | Boston | |
bordeaux.peerReviewed | oui | |
hal.identifier | hal-00771640 | |
hal.version | 1 | |
hal.invited | non | |
hal.proceedings | oui | |
hal.conference.organizer | IEEE | |
hal.popular | non | |
hal.audience | Internationale | |
hal.origin.link | https://hal.archives-ouvertes.fr//hal-00771640v1 | |
bordeaux.COinS | ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.date=2013&rft.au=BEAUMONT,%20Olivier&LARCHEV%C3%8AQUE,%20Hubert&MARCHAL,%20Loris&rft.genre=unknown |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |