The problem of scheduling real-time tasks with minimum jitter is particular
ly important in many control applications; nevertheless, it has rarely been
studied in the scientific literature. This paper presents a unconventional
scheduling approach for distributed static systems where tasks are periodi
c and have arbitrary deadlines, precedence, and exclusion constraints. The
solution presented in this work not only creates feasible schedules, but al
so minimizes jitter for periodic tasks. We present a general framework cons
isting of an abstract architecture model and a general programming model. W
e show how to design a surprisingly simple and flexible scheduling method b
ased on simulated annealing. Experimental results demonstrate the significa
nt improvement of our algorithm over earliest deadline first and rate monot
onic algorithms.