<p>In the rapidly evolving domain of cloud computing, serverless architectures have gained prominence for their ease of deployment and dynamic scalability to offer reliable and scalable options for a variety of Internet applications. Specifically, serverless application architectures are helpful to deploy diverse Internet-based applications such as latency sensitive web services, Internet of Things (IoT) use cases (e.g., smart healthcare, connected vehicles), and AI-driven services over cloud and edge computing platforms. In order to maximize a serverless application deployment benefits, it is necessary to dynamically orchestrate its services under important constraints such as functions’ dependencies, application computational requirements and security profiles. In existing literature, most of the works primarily addressed serverless application deployment issues over cloud platforms. However, serverless application deployment challenges related to edge computing environment are not addressed in important dimensions such as fair utilization of edge computational resources and offering a secure computing environment. In this work, we propose a novel algorithm called Security Aware Serverless Application Partitioning (SASAP, <a href="https://github.com/Chirag-Bhise/SASAP">https://github.com/Chirag-Bhise/SASAP</a>, 2025) to address key challenges such as ensuring application computational requirements and security requirements over both the edge and cloud environments. We also consider fair utilization of edge computational resources. In various test scenarios across scalable edge computing cluster sizes, it was observed that SASAP ensures 83.87% secured serverless application deployment and ensures 90.33% fulfillment of computational requirements. Whereas, the recent existing works ensure only 34.33% of secured deployment. Besides, SASAP is also ensuring fair utilization of the edge computing resources for serverless application deployment compared to the existing works.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Edge computational resources and security aware serverless application deployment

  • Chirag Anil Bhise,
  • Anil Kumar Rangisetti

摘要

In the rapidly evolving domain of cloud computing, serverless architectures have gained prominence for their ease of deployment and dynamic scalability to offer reliable and scalable options for a variety of Internet applications. Specifically, serverless application architectures are helpful to deploy diverse Internet-based applications such as latency sensitive web services, Internet of Things (IoT) use cases (e.g., smart healthcare, connected vehicles), and AI-driven services over cloud and edge computing platforms. In order to maximize a serverless application deployment benefits, it is necessary to dynamically orchestrate its services under important constraints such as functions’ dependencies, application computational requirements and security profiles. In existing literature, most of the works primarily addressed serverless application deployment issues over cloud platforms. However, serverless application deployment challenges related to edge computing environment are not addressed in important dimensions such as fair utilization of edge computational resources and offering a secure computing environment. In this work, we propose a novel algorithm called Security Aware Serverless Application Partitioning (SASAP, https://github.com/Chirag-Bhise/SASAP, 2025) to address key challenges such as ensuring application computational requirements and security requirements over both the edge and cloud environments. We also consider fair utilization of edge computational resources. In various test scenarios across scalable edge computing cluster sizes, it was observed that SASAP ensures 83.87% secured serverless application deployment and ensures 90.33% fulfillment of computational requirements. Whereas, the recent existing works ensure only 34.33% of secured deployment. Besides, SASAP is also ensuring fair utilization of the edge computing resources for serverless application deployment compared to the existing works.