ARTS: a distributed real-time kernel

Abstract

ARTS is a distributed real-time operating system designed for a real-time systems testbed being developed at Camegle Mellon University. The objective of the testbed is to develop and verify advanced real-time computing technologies for a distributed environment. The tastbed consists of a set of SUN3 workstations connected by a real-time network based on IEEE 802.5 Token Ring and Ethernet. The goal of the ARTS Kernel is not to produce simply a fast real-time executive, but rather to provide users with a predictable, analyzable, and reliable distributed real-time computing environment. In particular, we have developed a real-time object model which is incorporated with a time fence protocol. The time fence protocol is used at every invocation in the object to detect the origin of timing errors. We also developed an integrated time-driven scheduling model and its scheduler based on the notion of policy/mechanism separation. Since each scheduling policy is implemented as a kernel object, a user can easily add policies or change the system's scheduling policy. A real-time toolset was also developed in order to predict the schedulability of the real-time activities.In this paper, we give an overview of the ARTS Kernel and ARTS real-time toolset. In particular, we introduce a real-time object model and the integrated time-driven scheduling model. We then describe the basic primitives and major components of the ARTS Kernel and the real-time toolset which consists of the schedulability analyzer, Scheduler 1-2-3 , and the real-time monitor/debugger, ARM.