Online, non-preemptive scheduling of equal-length jobs on two identical machines

Abstract

We consider the non-preemptive scheduling of two identical machines for jobs with equal processing times yet arbitrary release dates and deadlines. Our objective is to maximize the number of jobs completed by their deadlines. Using standard nomenclature, this problem is denoted as PI |P j = p, r j |∑ Ū j. The problem is known to be polynomially solvable in an offline setting. In an online variant of the problem, a job's existence and parameters are revealed to the scheduler only upon that job's release date. We present an online, deterministic algorithm for the problem and prove that it is 3/2-competitive. A simple lower bound shows that this is the optimal deterministic competitiveness. © Springer-Verlag Berlin Heidelberg 2006.