Cloud Workflow Task and Virtualized Resource Collaborative Adaptive Scheduling Algorithm Based on Distributed Deep Learning

Cloud workflow task scheduling and resource allocation are core issues in cloud computing environment. Balancing the quality of user service and the revenue of service providers is a challenge in cloud workflow task scheduling and resource allocation. Therefore, we propose a cloud workflow task and virtualized resource collaborative adaptive scheduling algorithm based on heterogeneous distributed deep learning. This solves the multi-queue multi-cluster workflow task scheduling and resource allocation problem by combining multiple heterogeneous deep neural networks as the scheduling model of the cloud system. An optimal scheduling strategy is generated by minimizing the workflow task delay and energy consumption. Results from extensive experiments show that the proposed framework can effectively solve the multi-objective optimization problem of cloud Workflow task scheduling and resource allocation, and provide a nearly optimal scheduling strategy.

I. Introduction

The development of cloud computing has promoted the rapid development of the entire information industry. The computing power and storage capacity of the cloud computing platform can meet the needs of different users and provide high-quality customized services. The quality of a cloud platform's scheduling strategy determines its service quality and operating profit [1], so cloud Workflow task scheduling and resource allocation optimization have always been core issues.