WS‐DOA: Design of Hybrid Heuristic Algorithm for Deriving Multi‐Objective Function of Optimal Task Scheduling and VM Migration Over Cloud Sector
ABSTRACT Cloud‐based computing is an innovative computing model that utilizes a variety of self‐driving devices and adaptable computing structures. Efficient cloud computing relies on the critical step of scheduling tasks. In order to decrease energy use and increase service providers' profits...
Saved in:
| Published in: | Transactions on emerging telecommunications technologies Vol. 36; no. 4 |
|---|---|
| Main Authors: | , , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Chichester, UK
John Wiley & Sons, Ltd
01.04.2025
|
| Subjects: | |
| ISSN: | 2161-3915, 2161-3915 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | ABSTRACT
Cloud‐based computing is an innovative computing model that utilizes a variety of self‐driving devices and adaptable computing structures. Efficient cloud computing relies on the critical step of scheduling tasks. In order to decrease energy use and increase service providers' profits by speeding up processing, task planning remains crucial. Scheduling tasks represents one of the crucial operations of computing in the cloud. The main challenge in task scheduling is to allocate the complete task to a suitable Virtual Machine (VM) while ensuring profitability. Various scheduling techniques in the cloud ensure Quality of Service (QoS), but as task scaling increases, scheduling becomes more challenging. Hence, there is a need for enhanced scheduling. Previous studies did not cover task planning and VM migration, which effectively address resource utilization and energy efficiency. An advanced deep learning model with an enhanced heuristic algorithm is suggested to improve the scheduling process. This model aims to predict data that assist in task scheduling and VM migration through the derivation of a multi‐objective function. Initially, the cloud data are gathered from benchmark data sources. Further, the prediction is carried out by a Multiscale Dilated Recurrent Neural Network (MDRNN). To derive the multi‐objective function, the Water Strider‐based Dingo Optimization Algorithm (WS‐DOA) is proposed. Following the prediction, task scheduling is performed with the WS‐DOA to derive a multi‐objective function considering constraints like resource cost, energy consumption, response time, and security. Likewise, VM migration involves formulating the objective function with WS‐DOA, considering make span and cost. Finally, the proposed model is examined using diverse metrics. On the contrary, the enhanced method evinces that it acquires higher results for task scheduling and VM migration.
Initially, the cloud data are gathered from the benchmark data sources. Further, the prediction is carried out by using the Multiscale Dilated Recurrent Neural Network (MDRNN). To derive the multi‐objective function, the Water Strider‐based Dingo Optimization Algorithm (WS‐DOA) is proposed. Finally, the proposed model is examined using diverse metrics. On the contrary, the enhanced method evinces that it acquires higher results for task scheduling and VM migration. |
|---|---|
| ISSN: | 2161-3915 2161-3915 |
| DOI: | 10.1002/ett.70104 |