id: Simula.ND.325
ERROR IN ENCODING
title: Limits of work-stealing scheduling
publication_year: 2009
abstract: The number of applications with many parallel cooperating processes is   steadily increasing, and developing efficient runtimes for their execution   is an important task. Several frameworks have been developed, such as   MapReduce and Dryad, but developing scheduling mechanisms that take into   account processing \emph{and} communication requirements is hard.  In this   paper, we explore the limits of work stealing scheduler, which has   empirically been shown to perform well, and evaluate load-balancing based on   graph partitioning as an orthogonal approach.  All the algorithms are   implemented in our Nornir runtime system, and our experiments on a   multi-core workstation machine show that the main cause of performance   degradation of work stealing is when very little processing time, which we   quantify exactly, is performed per message.  This is the type of workload in   which graph partitioning has the potential to achieve better performance   than work-stealing.
publication_url: 
pdf_url: 
booktitle: Job Scheduling Strategies for Parallel Processing (14th International Workshop, 2009)
editor: Eitan Frachtenberg and Uwe Schwiegelshohn
volume: 
number: 
chapter: 
pages: 280-299
publisher: Springer Berlin / Heidelberg
organization: 
address: 
edition: 
series: 
isbn: 978-3-642-04632-2
keywords: ()
publication_month: 
note: 
annote: 
additional: []
location: 
publication_state: Published
simula_ou: [<Department at /simula/department/media>, <Department at /simula/research/mpg>]
publisher_url: 
category: Workshop
from_date: 2009/05/29 00:00:00 GMT+2
to_date: 2009/05/29 00:00:00 GMT+2