Title :
Apply an analytical approach for predictable real-time systems
Author :
Vogel, David ; Mesler, Thomas
Author_Institution :
IBM, Owego, NY, USA
Abstract :
The rate monotonic scheduling (RMS) theory in an analytical approach that solves the most fundamental problems in constructing real-time software: providing predictable real-time response in the face of a high processing load while still supporting modern development methodologies and ease of maintenance after product delivery. Additionally, when all time constraints cannot be met, the system must be able to shed load in a predictable manner. RMS theory assigns priorities to tasks based on their periodicity. Tasks with shorter periods receive the higher priorities. Using this analytical approach provides the predictability required of real-time systems while still supporting modern design techniques. Theorems for the RMS methodology are used to predict system behavior. There are three principal theorems used in RMS to determine the schedulability of a specified task set. A brief overview of the three RMS theorems is presented
Keywords :
real-time systems; scheduling; software engineering; analytical approach; design; predictable real-time systems; rate monotonic scheduling theory; real-time software; Displays; Process control; Processor scheduling; Real time systems; Time domain analysis; Time series analysis;
Conference_Titel :
Southern Tier Technical Conference, 1990., Proceedings of the 1990 IEEE
Conference_Location :
Binghamton, NY
DOI :
10.1109/STIER.1990.324632