i . ' il ' . Imprecise Results : Utilizing Partial Computations in Real - Time Systems
نویسندگان
چکیده
In real-time systems, a computation may not have time to complete its execution because of deadline requirements. In such cases. no result except the approximate results produced by the computation up to that point will be available. It is desirable to utilize these imprecise results if possible. We propose two approaches to enable computations to return imprecise results when executions cannot be completed normally. The milestone approach records results periodically. and if a deadline is reached, returns the last recorded result. The sieve approach demarcates sections of code which can be skipped if the time available is insufficient. By using these approaches, the system is able to produce imprecise results when deadlines are reached. We describe the design of the Concord project which supports imprecise computations using these techniques. We also present a general model of imprecise computations which takes into account the influence of the environment, and show where our approach fits into this model.
منابع مشابه
Genetic Algorithms in Real-Time Imprecise Computing
This article describes the use of genetic algorithms in real-time systems that employ the imprecise computation paradigm. In real-time systems, the focus is on ensuring that a set of tasks each complete within their deadlines. Faults may occur in the computation or the environment that can cause missed deadlines. That is why the idea of using partial results, when exact ones cannot be produced ...
متن کاملA Fast Algorithm for Scheduling Imprecise Computations with Timing Constraints to Minimize Weighted Error
Scheduling tasks with dflerent weights in the imprecise computation model is rather dflcult. Each task in the imprecise computation model is logically decomposed into a mandatory subtask and an optional subtask. The mandatory subtask must be completely executed before the deadline to produce acceptable result; the optional subtask begins after the mandatory subtask to refine the result. The err...
متن کاملEnhancing the Performance and Dependability of Real-time Systems
The imprecise-computation technique was proposed as a way to handle transient overloads and enhance the dependability of real-time systems. In a system based on this technique, each time-critical task produces an approximate but usable result whenever a failure or overload prevents the system from producing a precise result. This approach makes meeting deadlines easier, increases the availabili...
متن کاملMaximum Power Point Tracker for Photovoltaic Systems Based on Moth-Flame Optimization Considering Partial Shading Conditions
The performance of photovoltaic (PV) systems is highly dependent on environmental conditions. Due to probable changes in environmental conditions, the real-time control of PV systems is essential for exploiting their maximum possible power. This paper proposes a new method to track the maximum power point of PV systems using the moth-flame optimization algorithm. In this method, the PV DC-DC co...
متن کاملOperating System Support for Imprecise Computation
Recent research on imprecise computation (that is, flexible computation) at the University of Illinois has focused on the development of operating system support for imprecise computations. The scheduling algorithms and imprecise-computation environment resulting from this work enable the use of imprecise computation as a means to provide scheduling flexibility in real-time systems and to enhan...
متن کامل