Loading...

Workflow Ensembles Scheduling Improvement in Edge-Based Infrastructures

Nazari, Zahra | 2023

106 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 56400 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Habibi, Jafar
  7. Abstract:
  8. In the cloud computing environment, although many processing resources for application software with the volume of calculations The above is available, but it may not work for software that is sensitive to latency. In fact, due to the large distance between users and end devices to cloud servers, sending requests And the required data to the data centers requires a high bandwidth and therefore the data transfer time is also significant. In addition, cloud computing in platforms such as cloud-based IoT architecture faces other challenges such as connectivity, resource limitations of end devices, and security. In order to solve the challenges in the cloud computing environment and respond to the growing volume of delay-sensitive applications, the concept of edge computing has been proposed. In this type of infrastructure, computing resources are at the edge of the network and nearby Final devices are considered. Also, application software, often sensitive to delay and security, They have relatively less processing volume and high interaction of edge end devices with servers. On the other hand, achieving the goals and advantages of edge-based infrastructure requires the existence of codified and comprehensive management policies for existing resources, which itself faces many challenges. There are methods for displaying application software and considering the order of executive operations in them, one of which is workflow. Workflow can be used to model various software applications, including computational problems that involve intensive computations. In order to solve the mentioned challenges, we are trying to provide a solution to solve the task scheduling problem in the workflow groups in the edge-based infrastructure. The model of workflow groups is used in order to include the dependencies between tasks and also to determine the order of execution of existing tasks in applied software. In order to respond as best as possible to the needs of users, the implementation of each task in the presented model is considered to have two mandatory and optional parts, the mandatory part of a task must be fully implemented, and the goal will be to implement the optional part as much as possible according to the available resources and conditions. In this research, an algorithm for improving scheduling in workflow groups is presented
  9. Keywords:
  10. Resources Management ; Scheduling ; Edge Computing ; Task Scheduling Algorithm ; Mandatory Tasks ; Optional Tasks ; Workflow Ensemble

 Digital Object List