Title :
CAISARTS: a tool for real-time scheduling assistance
Author :
Humphrey, Marty ; Stankovic, John A.
Author_Institution :
Dept. of Comput. Sci., Massachusetts Univ., Amherst, MA, USA
Abstract :
CAISARTS (Conceptual, Analytical, and Implementation Scheduling Advice for Real-Time Systems) is a rule-based system used by real-time application designers to obtain expert assistance for all aspects of the design related to scheduling: granularity of tasks, allocation of tasks, choice and analysis of scheduling paradigm, analysis of overheads of particular operating systems and scheduling paradigms, and code templates for tasks. The rule base is partitioned; subsets of the rule base can be selected for firing, thus enabling the user to ask CAISARTS for advice and analysis relevant for different phases of the design. In contrast to existing real-time tools, CAISARTS attempts to cover the entire design process related to scheduling without focusing on, for example, solely schedulability analysis. A unique feature of CAISARTS is that its rule base is extensible by the user-a graphical interface is used to add new rules as new real-time results are identified. Challenges in the design of the initial rule set include how to design and partition the rule base so that it can be both easily modifiable and readily usable by the user in choosing rules to fire; how to encode rules that are inherently contradictory; how to encode ambiguous knowledge; and how to make the rules both comprehensive and precise. The effectiveness of CAISARTS is shown through its use on a representative distributed real-time system scenario with end-to-end constraints
Keywords :
intelligent design assistants; knowledge engineering; processor scheduling; real-time systems; software tools; CAISARTS; ambiguous knowledge; code templates; distributed real-time system scenario; end-to-end constraints; operating systems; real-time application designers; real-time scheduling assistance; rule-based system; scheduling paradigm; task allocation; task granularity; Application software; Computer science; Containers; Job shop scheduling; Knowledge based systems; Process design; Processor scheduling; Real time systems; Resource management; Scheduling algorithm;
Conference_Titel :
Real-Time Technology and Applications Symposium, 1996. Proceedings., 1996 IEEE
Conference_Location :
Brookline, MA
Print_ISBN :
0-8186-7448-2
DOI :
10.1109/RTTAS.1996.509532