Integrating Job Parallelism in Real-Time Scheduling Theory

TL;DR

This paper introduces a new task model with job parallelism for real-time scheduling on multiprocessors, providing complexity analysis, an optimal scheduling algorithm, and bounds on feasibility and migrations.

Contribution

It presents a novel task model integrating job parallelism and offers an optimal scheduling algorithm with feasibility bounds and migration control techniques.

Findings

Feasibility problem is linear in number of tasks for fixed processors.

Proposes a theoretically optimal scheduling algorithm.

Provides exact utilization bounds for feasibility.

Abstract

We investigate the global scheduling of sporadic, implicit deadline, real-time task systems on multiprocessor platforms. We provide a task model which integrates job parallelism. We prove that the time-complexity of the feasibility problem of these systems is linear relatively to the number of (sporadic) tasks for a fixed number of processors. We propose a scheduling algorithm theoretically optimal (i.e., preemptions and migrations neglected). Moreover, we provide an exact feasibility utilization bound. Lastly, we propose a technique to limit the number of migrations and preemptions.