Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Cloud Task Scheduling Using a Hybrid Particle Swarm and Grey Wolf Optimization Approach

Raveena Prasad, Aarush Roy, Suchi Kumari

TL;DR

This paper presents a new hybrid method that combines two popular algorithms, Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO), which offers strong global search capabilities (exploration), while PSO enhances local refinement (exploitation).

Abstract

Assigning tasks efficiently in cloud computing is a challenging problem and is considered an NP-hard problem. Many researchers have used metaheuristic algorithms to solve it, but these often struggle to handle dynamic workloads and explore all possible options effectively. Therefore, this paper presents a new hybrid method that combines two popular algorithms, Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO). GWO offers strong global search capabilities (exploration), while PSO enhances local refinement (exploitation). The hybrid approach, called HybridPSOGWO, is compared with other existing methods like MPSOSA, RL-GWO, CCGP, and HybridPSOMinMin, using key performance indicators such as makespan, throughput, and load balancing. We tested our approach using both a simulation tool (CloudSim Plus) and real-world data. The results show that HybridPSOGWO outperforms other methods, with up to 15\% improvement in makespan and 10\% better throughput, while also distributing tasks more evenly across virtual machines. Our implementation achieves consistent convergence within a few iterations, highlighting its potential for efficient and adaptive cloud scheduling.

Enhancing Cloud Task Scheduling Using a Hybrid Particle Swarm and Grey Wolf Optimization Approach

TL;DR

This paper presents a new hybrid method that combines two popular algorithms, Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO), which offers strong global search capabilities (exploration), while PSO enhances local refinement (exploitation).

Abstract

Assigning tasks efficiently in cloud computing is a challenging problem and is considered an NP-hard problem. Many researchers have used metaheuristic algorithms to solve it, but these often struggle to handle dynamic workloads and explore all possible options effectively. Therefore, this paper presents a new hybrid method that combines two popular algorithms, Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO). GWO offers strong global search capabilities (exploration), while PSO enhances local refinement (exploitation). The hybrid approach, called HybridPSOGWO, is compared with other existing methods like MPSOSA, RL-GWO, CCGP, and HybridPSOMinMin, using key performance indicators such as makespan, throughput, and load balancing. We tested our approach using both a simulation tool (CloudSim Plus) and real-world data. The results show that HybridPSOGWO outperforms other methods, with up to 15\% improvement in makespan and 10\% better throughput, while also distributing tasks more evenly across virtual machines. Our implementation achieves consistent convergence within a few iterations, highlighting its potential for efficient and adaptive cloud scheduling.

Paper Structure

This paper contains 24 sections, 1 equation, 5 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Metaheuristic Approaches for Task Scheduling
  4. Hybrid Metaheuristic Approaches
  5. Proposed Approach
  6. System Architecture
  7. Problem Formulation
  8. HybridPSOGWO Mechanism
  9. Adaptive Weight and Diversity Control Mechanisms
  10. Position Update Rule
  11. VM-Aware Task Mapping
  12. HybridPSOGWO Algorithm
  13. Results and Analysis
  14. Implementation Environment
  15. Performance Metrics
  16. ...and 9 more sections

Figures (5)

  • Figure 1: Architecture of the proposed HybridPSOGWO approach for cloud task scheduling, showing the integration of PSO and GWO components with VM-aware task mapping.
  • Figure 2: Detailed VM load analysis showing (a) individual VM task distribution and (b) load balance coefficients across algorithms on CloudSim Plus simulation.
  • Figure 3: Overall performance comparison of algorithms on CloudSim Plus simulation with 800 tasks and 4 VMs.
  • Figure 4: Real-world dataset performance analysis showing (a) combined metric scores and (b) task distribution heatmap on Google Borg traces with 800 tasks.
  • Figure 5: 3D Scalability analysis showing makespan and execution time scaling with increasing task counts.