Workshop on System Management Tools for Large-Scale Parallel Systems

Deadline is extended to Dec 15, 2006!

Third International Workshop on System Management Techniques, Processes, and Services (SMTPS)

held in conjunction with
The 2007 International Parallel and Distributed Processing Symposium
Long Beach, CA, March 30th, 2007

Workshop General Chair

Ramendra Sahoo
IBM Research
(rsahoo@us.ibm.com)

Technical Co-Chairs

Kyung Dong Ryu
IBM Research
(kryu@us.ibm.com)

Fabrizio Petrini
Pacific Northwest National Lab
(fabrizio.petrini@pnl.gov)

Yanyong Zhang
Rutgers University
(yyzhang@ece.rutgers.edu)

In today’s high-performance computing, system management and related services play a key role. With service business accounting for more than half of the U.S. economy, in our third year of SMTPS we would like to broaden the scope of our workshop to cover all aspects of system management, going beyond scientific computing. In order to satisfy the systems needs of both commercial and scientific applications, the focus on system management now includes not only the tools and user interfaces, but also other aspects such as services, processes and system control.

Businesses in the IT area are focusing more and more on innovative techniques, processes and methods to manage commercial or scientific systems remotely, in order to optimize the use of resources by minimizing system down time. As a result, there are requirements not only to revisit some of the traditional methods used to develop system management tools for today’s servers but also to evaluate the implications and benefits the new programming models can provide for parallel and distributed systems in terms of system services performance and utilization.

This workshop is intended to bring together researchers and practitioners to identify the new challenges imposed by this trend and investigating efficient software tools, techniques and service processes to improve the performance, reliability and operation of enterprise servers including parallel and distributed systems.

In addition to paper presentations by researchers in this area, we plan to invite industrial speakers to give their perspective on important research topics, and organizing a panel discussion on where future research is really needed.

Topics of interest include, and are not limited to :

Scalable operating system design
Scalable resource management tools
Efficient failure diagnosis, failure prediction and failure recovery tools
Scalable job scheduling tools
Scalable check-pointing tools
Self-healing and self-management tools
Power management for enterprise servers leading to efficient systems management
System bring-up and control tools
Ease of system maintenance, services including system management experiences
Performance, system utilization implications
Scalable I/O and file system management
Optimization techiniques for services management
Services engineering and utility computing techniques
Web services and Services oriented architecture and implications to system management aspects

Results of both theoretical and practical significance will be considered.

Call for papers: