<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
</head>
<body bgcolor="#ffffff" text="#000000">
<div class="moz-text-flowed"
style="font-family: -moz-fixed; font-size: 13px;" lang="x-western">Call
for Papers
<br>
<br>
---------------------------------------------------------------------------------------
<br>
The 2nd ACM Workshop on Many-Task Computing on Grids and Supercomputers
(MTAGS) 2009
<br>
<a class="moz-txt-link-freetext"
href="http://dsl.cs.uchicago.edu/MTAGS09/">http://dsl.cs.uchicago.edu/MTAGS09/</a>
<br>
---------------------------------------------------------------------------------------
<br>
November 16th, 2009
<br>
Portland, Oregon, USA
<br>
<br>
Co-located with with IEEE/ACM International Conference for
<br>
High Performance Computing, Networking, Storage and Analysis (SC09)
<br>
<br>
=======================================================================================
<br>
The 2nd workshop on Many-Task Computing on Grids and Supercomputers
(MTAGS) will
<br>
provide the scientific community a dedicated forum for presenting new
research,
<br>
development, and deployment efforts of loosely coupled large scale
applications on
<br>
large scale clusters, Grids, Supercomputers, and Cloud Computing
infrastructure.
<br>
Many-task computing (MTC), the theme of the workshop encompasses
loosely coupled
<br>
applications, which are generally composed of many tasks (both
independent and
<br>
dependent tasks) to achieve some larger application goal. This
workshop will cover
<br>
challenges that can hamper efficiency and utilization in running
applications on
<br>
large-scale systems, such as local resource manager scalability and
granularity,
<br>
efficient utilization of the raw hardware, parallel file system
contention and
<br>
scalability, reliability at scale, and application scalability. We
welcome paper
<br>
submissions on all topics related to MTC on large scale systems.
Papers will be
<br>
peer-reviewed, and accepted papers will be published in the workshop
proceedings as
<br>
part of the ACM digital library. The workshop will be co-located with
the
<br>
IEEE/ACM Supercomputing 2009 Conference in Portland Oregon on November
16th, 2009.
<br>
For more information, please visithttp://dsl.cs.uchicago.edu/MTAGS09/.
<br>
<br>
Scope
<br>
---------------------------------------------------------------------------------------
<br>
This workshop will focus on the ability to manage and execute large
scale applications
<br>
on today's largest clusters, Grids, and Supercomputers. Clusters with
50K+ processor
<br>
cores are beginning to come online (i.e. TACC Sun Constellation System
- Ranger), Grids
<br>
(i.e. TeraGrid) with a dozen sites and 100K+ processors, and
supercomputers with 160K
<br>
processors (i.e. IBM BlueGene/P). Large clusters and supercomputers
have traditionally
<br>
been high performance computing (HPC) systems, as they are efficient at
executing
<br>
tightly coupled parallel jobs within a particular machine with
low-latency
<br>
interconnects; the applications typically use message passing interface
(MPI) to
<br>
achieve the needed inter-process communication. On the other hand,
Grids have been the
<br>
preferred platform for more loosely coupled applications that tend to
be managed and
<br>
executed through workflow systems. In contrast to HPC (tightly coupled
applications),
<br>
these loosely coupled applications make up a new class of applications
as what we call
<br>
Many-Task Computing (MTC). MTC systems generally involve the execution
of independent,
<br>
sequential jobs that can be individually scheduled on many different
computing
<br>
resources across multiple administrative boundaries. MTC systems
typically achieve this
<br>
using various grid computing technologies and techniques, and often
times use files to
<br>
achieve the inter-process communication as alternative communication
mechanisms than
<br>
MPI. MTC is reminiscent to High Throughput Computing (HTC); however,
MTC differs from
<br>
HTC in the emphasis of using many computing resources over short
periods of time to
<br>
accomplish many computational tasks, where the primary metrics are
measured in seconds
<br>
(e.g. FLOPS, tasks/sec, MB/s I/O rates). HTC on the other hand requires
large amounts
<br>
of computing for longer times (months and years, rather than hours and
days, and are
<br>
generally measured in operations per month).
<br>
<br>
Today's existing HPC systems are a viable platform to host MTC
applications. However,
<br>
some challenges arise in large scale applications when run on large
scale systems,
<br>
which can hamper the efficiency and utilization of these large scale
systems. These
<br>
challenges vary from local resource manager scalability and
granularity, efficient
<br>
utilization of the raw hardware, shared file system contention and
scalability,
<br>
reliability at scale, application scalability, and understanding the
limitations of the
<br>
HPC systems in order to identify good candidate MTC applications.
Furthermore, the MTC
<br>
paradigm can be naturally applied to the emerging Cloud Computing
paradigm due to its
<br>
loosely coupled nature, which is being adopted by industry as the next
wave of
<br>
technological advancement to reduce operational costs while improving
efficiencies in
<br>
large scale infrastructures.
<br>
<br>
For an interesting discussion in a blog by Ian Foster on the difference
between MTC and
<br>
HTC, please see his blog
athttp://ianfoster.typepad.com/blog/2008/07/many-tasks-comp.html.
<br>
We also published two papers that are highly relevant to this workshop.
One paper is
<br>
titled "Toward Loosely Coupled Programming on Petascale Systems", and
was published in
<br>
SC08; the second paper is titled "Many-Task Computing for Grids and
Supercomputers",
<br>
which was published in MTAGS08. Furthermore, to see last year's
workshop program agenda,
<br>
and accepted papers and presentations, please
seehttp://dsl.cs.uchicago.edu/MTAGS08/.
<br>
For more information, please visithttp://dsl.cs.uchicago.edu/MTAGS09/.
<br>
<br>
Topics
<br>
---------------------------------------------------------------------------------------
<br>
MTAGS 2008 topics of interest include, but are not limited to:
<br>
* Compute Resource Management in large scale clusters, large Grids,
Supercomputers,
<br>
or Cloud Computing infrastructure
<br>
o Scheduling
<br>
o Job execution frameworks
<br>
o Local resource manager extensions
<br>
o Performance evaluation of resource managers in use on large
scale systems
<br>
o Challenges and opportunities in running many-task workloads on
HPC systems
<br>
o Challenges and opportunities in running many-task workloads
<br>
on Cloud Computing infrastructure
<br>
* Data Management in large scale Grid and Supercomputer
environments:
<br>
o Data-Aware Scheduling
<br>
o Parallel File System performance and scalability in large
deployments
<br>
o Distributed file systems
<br>
o Data caching frameworks and techniques
<br>
* Large-Scale Workflow Systems
<br>
o Workflow system performance and scalability analysis
<br>
o Scalability of workflow systems
<br>
o Workflow infrastructure and e-Science middleware
<br>
o Programming Paradigms and Models
<br>
* Large-Scale Many-Task Applications
<br>
o Large-scale many-task applications
<br>
o Large-scale many-task data-intensive applications
<br>
o Large-scale high throughput computing (HTC) applications
<br>
o Quasi-supercomputing applications, deployments, and
experiences
<br>
<br>
Paper Submission and Publication
<br>
---------------------------------------------------------------------------------------
<br>
Authors are invited to submit papers with unpublished, original work of
not more than
<br>
10 pages of double column text using single spaced 10 point size on 8.5
x 11 inch pages,
<br>
as per ACM 8.5 x 11 manuscript guidelines
<br>
(<a class="moz-txt-link-freetext"
href="http://www.acm.org/publications/instructions_for_proceedings_volumes">http://www.acm.org/publications/instructions_for_proceedings_volumes</a>);
document
<br>
templates can be found
athttp://www.acm.org/sigs/publications/proceedings-templates.
<br>
A 250 word abstract (PDF format) must be submitted online at
<br>
<a class="moz-txt-link-freetext"
href="https://cmt.research.microsoft.com/MTAGS2009/">https://cmt.research.microsoft.com/MTAGS2009/</a>
before the deadline of August 1st, 2009
<br>
at 11:59PM PST; the final 10 page papers in PDF format will be due on
September 1st,
<br>
2009 at 11:59PM PST. Papers will be peer-reviewed, and accepted papers
will be
<br>
published in the workshop proceedings as part of the ACM digital
library. Notifications
<br>
of the paper decisions will be sent out by October 1st, 2009. Selected
excellent work
<br>
will be invited to submit extended versions of the workshop paper to
the
<br>
IEEE Transactions on Parallel and Distributed Systems (TPDS) Journal,
Special Issue on
<br>
Many-Task Computing (due December 21st, 2009); for more information
about this journal
<br>
special issue, please visithttp://dsl.cs.uchicago.edu/TPDS_MTC/.
Submission implies
<br>
the willingness of at least one of the authors to register and present
the paper. For
<br>
more information, please visithttp://dsl.cs.uchicago.edu/MTAGS09/.
<br>
<br>
Important Dates
<br>
---------------------------------------------------------------------------------------
<br>
* Abstract Due: August 1st, 2009
<br>
* Papers Due: September 1st, 2009
<br>
* Notification of Acceptance: October 1st, 2009
<br>
* Camera Ready Papers Due: November 1st, 2009
<br>
* Workshop Date: November 16th, 2009
<br>
<br>
<br>
<br>
Committee Members
<br>
---------------------------------------------------------------------------------------
<br>
Workshop Chairs
<br>
* Ioan Raicu, University of Chicago
<br>
* Ian Foster, University of Chicago& Argonne National
Laboratory
<br>
* Yong Zhao, Microsoft
<br>
<br>
Technical Committee (confirmed)
<br>
* David Abramson, Monash University, Australia
<br>
* Pete Beckman, Argonne National Laboratory, USA
<br>
* Peter Dinda, Northwestern University, USA
<br>
* Ian Foster, University of Chicago& Argonne National
Laboratory, USA
<br>
* Bob Grossman, University of Illinois at Chicago, USA
<br>
* Indranil Gupta, University of Illinois at Urbana Champaign, USA
<br>
* Alexandru Iosup, Delft University of Technology, Netherlands
<br>
* Kamil Iskra, Argonne National Laboratory, USA
<br>
* Chuang Liu, Ask.com, USA
<br>
* Zhou Lei, Shanghai University, China
<br>
* Shiyong Lu, Wayne State University, USA
<br>
* Reagan Moore, University of North Carolina at Chapel Hill, USA
<br>
* Marlon Pierce, Indiana University, USA
<br>
* Ioan Raicu, University of Chicago, USA
<br>
* Matei Ripeanu, University of British Columbia, Canada
<br>
* David Swanson, University of Nebraska, USA
<br>
* Greg Thain, Univeristy of Wisconsin, USA
<br>
* Matthew Woitaszek, The University Corporation for Atmospheric
Research, USA
<br>
* Mike Wilde, University of Chicago& Argonne National
Laboratory, USA
<br>
* Sherali Zeadally, University of the District of Columbia, USA
<br>
* Yong Zhao, Microsoft, USA
<br>
<br>
<br>
<br>
</div>
</body>
</html>