Brief Description of Project:
Future applications will run on large-scale systems characterized by thousands of platforms, sensors,
decision nodes, actuators, and operators connected through heterogeneous networks to achieve their objectives.
The networks, operating systems, middleware, and applications that populate these systems offer a potentially
combinatoric number of configuration points for adjusting their resource requirements and the quality of service
(QoS) they deliver. In recent years, standards-based architectures and platforms (such as the OMG Data
Distribution Service and CORBA's Real-time Notification Service) have emerged that enable information management
systems to communicate by publishing the information they have and subscribing to the information they need.
The key to the success of these approaches is their ability to specify and enforce performance requirements between
different parts of information management systems using QoS parameters that (1) configure the networks, operating
systems, and middleware and (2) establish contracts that precisely specify a wide variety of QoS properties.
This project will focus on the R&D challenges associated with the following topics pertaining to evaluating architectures,
platforms, and standards that support information management in large-scale distributed systems:
-
Approaches for integrating real-time pub/sub platforms over network
transports, such as UDP, multicast, and/or wireless networks.
-
Approaches for benchmarking standards- and COTS-based architectures
that will not only compare different implementations of the same
pub/sub technology, but also pinpoint use cases and environments in
which one technology is more well-suited than another.
|
