14th WORKSHOP ON JOB SCHEDULING STRATEGIES FOR PARALLEL PROCESSING

The JSSPP workshop, regularly ranked in the top 10% of Citeseer's
impact lists, addresses most aspects of parallel scheduling. More
than ever, this topic has grown in relevance and scope, and includes
not only traditional supercomputers and clusters, but also emerging
platforms and paradigms, such as multi-core and many-core systems,
virtualization, Grids, and cloud computing. Accordingly, JSSPP
solicits papers that fall under but are not limited to any of the
following themes:

* Experience with and design of parallel scheduling systems of all
scales.
* Performance evaluation, methodology, and simulation of parallel
scheduling, including benchmarks and metrics.
* Workload characterization, classification, and modeling.
* General scheduling aspects like fairness, priorities, accounting
issues, performance guarantees, QoS, and load estimation.
* Scheduling within virtualized systems, Grids, compute clouds,
including interaction of local schedulers and higher level
schedulers.
* Effect of scheduling on application performance, interactivity,
and interaction with memory, caches, and I/O performance.


Submission dates and guidelines
===============================

DEADLINE: February 13, 2009
NOTIFICATION: March 16, 2009
FINAL PAPER DUE: April 10, 2009

Papers should be no longer than 20 single-spaced pages, 10pt font,
including figures and references. All papers in scope will be
reviewed. As accepted final papers will have to follow the LNCS format,
authors may choose to already pre-format their submittal based on the
instructions at Springer's web site:

http://www.springer.com/computer/lncs?SGWID=0-164-7-72376-0

Files should be submitted in PDF format. Authors must ensure that
electronically submitted files are formatted for 8.5x11 inch paper.
Submission procedures will be available six weeks before submission
deadline at the workshop's homepage:

http://www.cs.huji.ac.il/~feit/parsched/


Goals
=====

Continuing the tradition established at IPPS'95, the workshop is
intended to attract people from academia, industry, supercomputing
centers, national laboratories, Grid initiatives, and parallel
computer vendors to address resource management issues in parallel
systems, and attempt to resolve conflicting goals such as short
response times for interactive work, minimal interference with batch
jobs, fairness to users based on their priorities, and high system
utilization. We aim to balance between
* reports of current practices in the entire range of parallel
systems,
* proposals of novel schemes that have not yet been tested in a
real environment, and
* realistic models and their analysis.


Registration
============

Registration will be part of the IPDPS process and is handled by the IEEE.
For details, see the IPDPS web site

http://www.ipdps.org/


Proceedings
===========

Interim proceedings containing a collection of the papers presented
will be distributed at the workshop. It is planned to also publish a
post-workshop proceedings in the Springer-Verlag Lecture Notes on
Computer Science series, as was done in previous years (pending
approval from Springer).


Workshop organizers
===================

Eitan Frachtenberg, Powerset
Uwe Schwiegelshohn, Technische Universität Dortmund


Program Committee
=================

Su-Hui Chiang, Portland State University
Walfredo Cirne, Google
Allen Downey, Olin College
Dror Feitelson, The Hebrew University
Allan Gottlieb, New York University
Andrew Grimshaw, University of Virginia
Moe Jette, Lawrence Livermore National Lab
Virginia Lo, University of Oregon
Jose Moreira, IBM Watson Research Center
Reagan Moore, San Diego Supercomputer Center
Bill Nitzberg, Altair Engineering
Mark Squillante, IBM Watson Research Center
Dan Tsafrir, IBM Watson Research Center
John Towns, National Center for Supercomputing Applications
Jon Weissman, University of Minnesota
Ramin Yahyapour, Technische Universität Dortmund