Call For Papers

AOHPS: 1st Workshop on
Application-Oriented High-Performance Systems

To be held in conjunction with the 13th International Symposium on High-Performance Computer Architecture (HPCA-13)

Hyatt Regency Phoenix - Phoenix, AZ - February 10-14, 2007

Workshop Chair:
William R. Dieter, Univ. of Kentucky

Program Committee:
Pavan Balaji, Argonne National Lab
Hank Dietz, Univ. of Kentucky
Wu Feng, Virginia Tech.
Randy Fisher, Aggregate.Org
Others TBD

Important Dates:
Abstracts Due12/08/06
Full Paper Due12/15/06
Notification of Acceptance01/05/06
Camera Ready Copy Due01/19/06

For more information send email to: dieter@engr.uky.edu

Application-oriented systems are designed to meet the needs of a single application or a small set of applications. Recently, high-performance computing technologies ranging from cluster computing to FPGAs have made it relatively easy to tune the design of a system for a particular application. When a computer system will be used primarily for a single application or a small set of applications, optimizing it for that workload often yields much shorter execution time, lower purchase price, less power consumption, smaller size, and/or lower operating cost.

Component prices have fallen far enough that application-optimized systems containing hundreds or thousands of components can be purchased with a modest budget. While such a large-scale design may yield the best performance for a given purchase price, component replacement, power consumption, cooling, and system maintenance costs can add up to a large percentage of the purchase price.

This workshop solicits papers in areas related to design, implementation, and experience with high-performance application-oriented computer systems. Topics of interest include, but are not limited to:

  • Application-aware design of parallel architecture
  • Tool and models for application-oriented cluster design
  • System life cycle, reliability, and maintenance
  • Tools, performance models, and experiences toward turnkey cluster applications
  • Practice and experience with application-specific computer system design
  • Cluster-in-context, environmentally-aware design
  • Life-cycle issues, TCO & maintenance, packaging and power/cooling issues
  • Application-oriented models and tools for design implementation, and operation of embedded systems