Schedulability Analysis of Global Scheduling Algorithms on Multiprocessor Platforms

Abstract

Abstract-This paper addresses the schedulability problem of periodic and sporadic real-time task sets with constrained deadlines preemptively scheduled on a multiprocessor platform composed by identical processors. We assume that a global work-conserving scheduler is used and migration from one processor to another is allowed during task lifetime. First, a general method to derive schedulability conditions for multiprocessor real-time systems will be presented. The analysis will be applied to two typical scheduling algorithms: Earliest Deadline First (EDF) and Fixed Priority (FP). Then, the derived schedulability conditions will be tightened, refining the analysis with a simple and effective technique that significantly improves the percentage of accepted task sets. The effectiveness of the proposed test is shown through an extensive set of synthetic experiments.

Multiprocessor extensions to real-time calculus

“…In some aspects, the presented analysis is also similar to results by Bertogna et al [2], Shin et al [10], and Zhang and Burns [12]. The main difference between our work and these prior efforts is that we consider more general task arrival and execution models.…”

“…By (3), we have E i (max(0, α + i (X))) ≤ e i · (max(0, α + i (X))) + v i . By (2), e i · (max(0, α + i (X))) + v i ≤ e i · (max(0, R i · X + B i )) + v i = max(0, e i · R i · X + e i · B i ) + v i . By Def.…”

“…7 is similar to GEDF. A number of GEDF schedulability tests have been developed assuming that jobs arrive periodically or sporadically (e.g., [1,2,6]). In this paper, we extend techniques from [1] and [6] in order to incorporate more general job arrivals and execution models.…”

A hierarchical multiprocessor bandwidth reservation scheme with timing guarantees

“…Note that, if all M s ≥ 2 processors are fully available, then a HRT GEDF schedulability test (e.g., (Baruah, 2007;Bertogna et al, 2009;Baruah and Baker, 2008)) can be applied to τ before calculating tardiness bounds. If this test passes, then maximum tardiness is zero.…”

FPSL, FPCL and FPZL schedulability analysis

“…Baker (2003) developed a fundamental schedulability test strategy, based on considering the minimum amount of interference in a given interval that is necessary to cause a deadline to be missed, and then taking the contra-positive of this to form a sufficient schedulability test. This basic strategy underpins an extensive thread of subsequent research into schedulability tests for global EDF Bertogna, 2007;Baruah and Baker, 2009;, global FP (Baruah and Fisher, 2008;Bertogna et al, 2009;Baker, 2006;Fisher and Baruah, 2006), and EDZL scheduling (Cirinei and Baker, 2007). …”