Algorithms for measuring the loads on the communication links in a distributed system are presented. Timestamped messages are used to measure latencies across physical and virtual links in the system. The algorithms do not generate any additional message traffic. Instead, load measurement information is opportunistically piggybacked onto application messages. The link load values can be used to initiate process migration and for dynamic flow control. An example of how loads on virtual links may be measured is given.>

Worst-case scheduling techniques for real-time applications often result in sever underutilization of the processor resources since most tasks finish in much less time than their anticipated worst-case execution times. A description is presented of compiler-based techniques that classify the application code on the basis of predictability and monotonicity, introduce measurement code fragments at selected points in the application code, and use the results of run-time measurements to dynamically adapt worst-case schedules. This results in better utilization of the system and early failure detection and recovery.>

Robot applications, represented as plans, are used to outline a viewpoint that robustness needs to be emphasized in two areas: in the plan representation and in the underlying system software. Robot applications are inherently distributed, since the hardware usually comprises a set of independent actuators and sensors, with the robot programs acting as links between them. A special model of distributed computation, the RS (Robot Schemas) model, has been designed to handle the issues of robot plan representation, and an overview of the model is presented. An initial implementation of the model with minimal execution support demonstrated that the domain-dependent aspect of robustness on its own was not sufficient for robust behavior. Consequently, the OS has been augmented with real-time scheduling, and monitoring facilities.>

An examination is made of the ability of the object-oriented model to represent and encapsulate the temporal characteristics of adaptable, real-time software. The differences between object-oriented design and the more traditional function-oriented design of real-time systems are briefly reviewed. The various temporal characteristics present in a real-time, object-oriented system and techniques for the design of object-oriented real-time software are examined. The many open problems in the area are discussed. The object-oriented model and many of the concepts described have been used in the implementation of the control software for an experimental robotics project.>

We describe our experiences with a real-time multiprocessor operating system, called GEM (Generalized Executive for Multiprocessor applications) and with an extension to GEM, called CHAOS (Concurrent Hierarchical Adaptable Object System). CHAOS offers kernel-level primitives that allow high-performance, large-scale, real-time software to be programmed as a system of interacting objects. This significantly improves modularity, reconfigurability, and maintainability.