Production printing workflow modeling and scheduling
Enable collective intelligence among multiple resources to reduce cost
and support value-added services
Rutgers University
Investigated a programming framework (named
Accord)
to support programming and runtime management of adaptive Grid applications.
The framework enabled dynamic composition of individual services, and
runtime functional / performance based adaptation of distributed
applications.
Accord is a sub project of umbrella project
AutoMate, which
investigates key technologies to enable the development of autonomic Grid
applications that are context aware and are capable of self-management.
Researched and developed a distributed interactive computational steering
system (named
DIOS++) to enable dynamical rule-based management of distributed &
parallel applications.
DIOS++ forms the back-end of
DISCOVER
Sandia National Labs
Introduced self-optimizing and self-healing features into DoE (Dept. of
Energy) CCA (Common
Component Architecture) CCAFFEINE framework to improve the performance of
parallel and distributed scientific applications. Enabled the adaptive
computational behaviors of components, and dynamic replacement of components
based on user-defined rules at runtime.