[Paper Review] Multiprocessor Scheduling For Tasks With Priority Using GA

G. Padmavathi, S. R. Vijayalakshmi|arXiv (Cornell University)|Jan 12, 2010

Distributed and Parallel Computing Systems3 citations

TL;DR

This paper proposes a genetic algorithm (GA)-based approach to optimize multiprocessor scheduling for priority-based real-time tasks, aiming to minimize makespan. By encoding task-to-processor assignments as chromosomes and evolving them via selection, crossover, and mutation, the GA efficiently explores the search space to achieve near-optimal scheduling with improved performance over traditional methods.

ABSTRACT

Multiprocessors have emerged as a powerful computing means for running realtime applications, especially where a uniprocessor system would not be sufficient enough to execute all the tasks. The high performance and reliability of multiprocessors have made them a powerful computing resource. Such computing environment requires an efficient algorithm to determine when and on which processor a given task should execute. In multiprocessor systems, an efficient scheduling of a parallel program onto the processors that minimizes the entire execution time is vital for achieving a high performance. This scheduling problem is known to be NPHard. In multiprocessor scheduling problem, a given program is to be scheduled in a given multiprocessor system such that the programs execution time is minimized. The last job must be completed as early as possible. Genetic algorithm (GA) is one of the widely used techniques for constrained optimization problems. Genetic algorithms are basically search algorithms based on the mechanics of natural selection and natural genesis. The main goal behind research on genetic algorithms is robustness i.e. balance between efficiency and efficacy. This paper proposes Genetic algorithm to solve scheduling problem of multiprocessors that minimizes the make span.

Motivation & Objective

• To address the NP-hard multiprocessor scheduling problem in real-time systems where tasks have priority constraints.
• To develop an efficient scheduling algorithm that minimizes the total execution time (makespan) in multiprocessor environments.
• To leverage genetic algorithms for robust optimization in complex, constrained scheduling scenarios.
• To improve system performance and reliability by reducing task completion time through intelligent processor assignment.
• To demonstrate the effectiveness of GA in handling priority-based task scheduling with high computational complexity.

Proposed method

• The scheduling problem is modeled as a combinatorial optimization task with priority constraints on tasks.
• A genetic algorithm is employed, encoding task-to-processor assignments as binary or integer chromosomes.
• Selection, crossover, and mutation operators are applied to evolve the population toward lower makespan.
• Fitness function is defined as the inverse of makespan, favoring shorter completion times.
• The algorithm iteratively evolves solutions over generations to converge on high-quality schedules.
• The approach handles precedence and priority constraints by incorporating them into the fitness evaluation and repair mechanisms.

Experimental results

Research questions

• RQ1 Can a genetic algorithm effectively minimize makespan in multiprocessor scheduling with priority-constrained tasks?
• RQ2 How does the proposed GA compare to conventional scheduling heuristics in terms of makespan and convergence?
• RQ3 To what extent can GA balance exploration and exploitation in the search space of NP-hard scheduling problems?
• RQ4 How well does the GA preserve task priority and precedence constraints during evolution?
• RQ5 What is the scalability and robustness of the GA approach across different task sets and processor counts?

Key findings

• The proposed GA approach achieves a significant reduction in makespan compared to traditional heuristic methods.
• The algorithm demonstrates robust convergence across multiple test cases with varying task priorities and workloads.
• Fitness evaluation based on makespan effectively guides the search toward optimal or near-optimal solutions.
• The inclusion of priority constraints in the fitness function ensures that high-priority tasks are scheduled earlier without violating dependencies.
• The method shows scalability and stability in handling diverse task sets and processor configurations.
• The results confirm that GA is a viable and effective approach for solving NP-hard multiprocessor scheduling problems with priority constraints.