Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimizing Task Scheduling in Heterogeneous Computing Environments: A Comparative Analysis of CPU, GPU, and ASIC Platforms Using E2C Simulator

Ali Mohammadjafari, Poorya Khajouie

TL;DR

This paper tackles the challenge of efficient task scheduling in heterogeneous computing environments comprising CPU, GPU, and ASIC platforms. It adopts a systematic benchmarking approach using the E2C simulator to evaluate four scheduling algorithms—FCFS, FCFS-NQ, MECT, and MEET—across low, medium, and high workload scenarios. The study finds that FCFS variants struggle under higher load, while MECT and MEET provide robust throughput and energy efficiency, with MEET prioritizing shorter execution times and MECT focusing on minimizing completion time. The resulting insights inform resource allocation decisions aimed at reducing task completion times and energy consumption in heterogeneous systems, with practical relevance for energy-constrained data centers and edge-to-cloud deployments.

Abstract

Efficient task scheduling in heterogeneous computing environments is imperative for optimizing resource utilization and minimizing task completion times. In this study, we conducted a comprehensive benchmarking analysis to evaluate the performance of four scheduling algorithms First Come, First-Served (FCFS), FCFS with No Queuing (FCFS-NQ), Minimum Expected Completion Time (MECT), and Minimum Expected Execution Time (MEET) across varying workload scenarios. We defined three workload scenarios: low, medium, and high, each representing different levels of computational demands. Through rigorous experimentation and analysis, we assessed the effectiveness of each algorithm in terms of total completion percentage, energy consumption, wasted energy, and energy per completion. Our findings highlight the strengths and limitations of each algorithm, with MECT and MEET emerging as robust contenders, dynamically prioritizing tasks based on comprehensive estimates of completion and execution times. Furthermore, MECT and MEET exhibit superior energy efficiency compared to FCFS and FCFS-NQ, underscoring their suitability for resource-constrained environments. This study provides valuable insights into the efficacy of task scheduling algorithms in heterogeneous computing environments, enabling informed decision-making to enhance resource allocation, minimize task completion times, and improve energy efficiency

Optimizing Task Scheduling in Heterogeneous Computing Environments: A Comparative Analysis of CPU, GPU, and ASIC Platforms Using E2C Simulator

TL;DR

This paper tackles the challenge of efficient task scheduling in heterogeneous computing environments comprising CPU, GPU, and ASIC platforms. It adopts a systematic benchmarking approach using the E2C simulator to evaluate four scheduling algorithms—FCFS, FCFS-NQ, MECT, and MEET—across low, medium, and high workload scenarios. The study finds that FCFS variants struggle under higher load, while MECT and MEET provide robust throughput and energy efficiency, with MEET prioritizing shorter execution times and MECT focusing on minimizing completion time. The resulting insights inform resource allocation decisions aimed at reducing task completion times and energy consumption in heterogeneous systems, with practical relevance for energy-constrained data centers and edge-to-cloud deployments.

Abstract

Efficient task scheduling in heterogeneous computing environments is imperative for optimizing resource utilization and minimizing task completion times. In this study, we conducted a comprehensive benchmarking analysis to evaluate the performance of four scheduling algorithms First Come, First-Served (FCFS), FCFS with No Queuing (FCFS-NQ), Minimum Expected Completion Time (MECT), and Minimum Expected Execution Time (MEET) across varying workload scenarios. We defined three workload scenarios: low, medium, and high, each representing different levels of computational demands. Through rigorous experimentation and analysis, we assessed the effectiveness of each algorithm in terms of total completion percentage, energy consumption, wasted energy, and energy per completion. Our findings highlight the strengths and limitations of each algorithm, with MECT and MEET emerging as robust contenders, dynamically prioritizing tasks based on comprehensive estimates of completion and execution times. Furthermore, MECT and MEET exhibit superior energy efficiency compared to FCFS and FCFS-NQ, underscoring their suitability for resource-constrained environments. This study provides valuable insights into the efficacy of task scheduling algorithms in heterogeneous computing environments, enabling informed decision-making to enhance resource allocation, minimize task completion times, and improve energy efficiency
Paper Structure (26 sections, 7 figures, 4 tables)

This paper contains 26 sections, 7 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Scheduling
  3. Understanding the Significance of Process Scheduling Optimization
  4. E2C: A Visual Simulator for Heterogeneous Computing Systems
  5. Proposed Approach and Research Framework
  6. Related Works
  7. Benchmarking Task Scheduling Algorithms
  8. Machine Learning Approaches:
  9. Adaptive Scheduling Strategies:
  10. Real-Time Scheduling
  11. Methodology
  12. E2C Simulator
  13. Simulation Components and Features
  14. Workload Generation and Task Specifications
  15. Scheduling Algorithms
  16. ...and 11 more sections

Figures (7)

  • Figure 1: Different types of CPU Scheduling Algorithms
  • Figure 2: Summary of the E2C Simulator comprising essential elements, namely the input workload, a queue for incoming tasks, a scheduler (also known as a load balancer), and a variety of machines depicted in distinct hues mokhtari2023e2c.
  • Figure 3: Total Completion Percentage.
  • Figure 4: Total Wasted Energy.
  • Figure 5: Total Consumed Energy.
  • ...and 2 more figures