An intelligent secure and efficient workflow scheduling (SEWS) model for heterogeneous cloud computing environment

An intelligent secure and efficient workflow scheduling (SEWS) model for heterogeneous cloud computing environment

연구 분야: Cryptography

학회: Knowledge and Information Systems

This study recognizes the critical role of the cloud computing platform in scientific workflow applications yet identifies vulnerabilities in existing cloud workflow systems, such as information leaks, unauthorized access, and compromised data integrity during task scheduling. Mainly, attackers exploit the lack of security for intermediate-level task information. To address these security threats, this work introduces the secure and efficient workflow scheduling (SEWS) model for heterogeneous cloud computing environments. The SEWS model identifies malicious attacks on all workflow tasks and focuses explicitly on safeguarding intermediary data. The SEWS model employs intelligent techniques to enhance security and introduces a comprehensive metric to measure the security of workflow tasks, considering factors like integrity, confidentiality, and availability. Beyond security improvements, the SEWS model aims to elevate the overall quality of service (QoS) in workflow scheduling applications. This includes reducing simulation time, enhancing overall power efficiency, and minimizing average energy consumption. Results: Results from the SEWS model demonstrate substantial improvements over the energy-minimized scheduling (EMS) model, with a reduction of 79.41% in average simulation time, 87.92% in average power sum, 41.35% in average power average, and 89.62% in average energy consumption. These findings underscore the SEWS model’s effectiveness in providing enhanced security and improved QoS in cloud workflow scheduling. The overarching goal of this work is to contribute to developing a more secure and efficient cloud workflow scheduling system, aligning with the increasing demands for robust security measures and optimized performance in heterogeneous cloud environments. Findings: Compared to the energy-minimized scheduling (EMS) model, the findings of this study demonstrate that the secure and efficient workflow scheduling (SEWS) model yields superior outcomes across key performance metrics. Specifically, the SEWS model excels in average simulation time, power sum, power average, and energy consumption. These results underscore the effectiveness of the SEWS model in enhancing the efficiency and resource utilization of cloud workflow scheduling. Importantly, the study identifies a notable gap in the existing work related to workflow task scheduling. Many prior studies still need to address the critical aspects of security and QoS in this context. While some jobs have attempted to enhance security, a significant limitation is the failure to extend these security measures to intermediary data. This gap in the literature highlights the unique contribution of the SEWS model, which addresses security concerns comprehensively and prioritizes QoS in the workflow task scheduling process. The observed superiority of the SEWS model in comparison with the EMS model serves as a testament to the model’s efficacy in concurrently addressing security and QoS challenges. By focusing on intermediary data, the SEWS model presents a holistic solution that aligns with the increasing demand for comprehensive security measures in cloud workflow environments. The findings emphasize the significance of integrating security and QoS considerations to establish a more robust and efficient workflow scheduling framework in heterogeneous cloud computing environments.