Specification for a Real-Time Transport Protocol

Abstract

Conventional standard transport protocols do not adequately support distributed real-time systems. An ideal protocol would ensure that at all times the communications subsystem is working on the task (message) of the most value to the overall system. In this paper we investigate how to develop a flexible and potent discrimination scheme based on importance, an abstract concept that measures the degree to which a task contributes to the overall system goal. We examine typical communication subsystems for what can and can not be expected from such a policy. Finally, we review our chain of reasoning in reaching a proposal for a scheme implementable within currently existing or developing transport protocols. Note: Abstract extracted from PDF text