Muhammad FahadFollow

Degree

Master of Science in Computer Science

Department

Department of Computer Science

School

School of Mathematics and Computer Science (SMCS)

Date of Submission

Fall 2023

Supervisor

Dr. Imran Khan, Assistant Professor, Department of Computer Science

Committee Member 1

Dr. Sajjad Haider, Examiner - I, Supervisor, Department of Computer Science, Institute of Business Administration, Karachi

Committee Member 2

Farhan Ahmed Siddiqui, Examiner - II, Department of Computer Science, Institute of Business Administration, Karachi

Keywords

Design Science Research, Vendor Specific Services, Function-as-a-Service, serverless, Cloud-Migration

Abstract

Serverless function or Function as a Service (FaaS) is a service delivery model derived from Platform as a Service (PaaS). Each Cloud Service Provider (CSP) offers its own flavor of FaaS keeping the underline concept the same which is to abstract away the implementation and infrastructure level details of the platform from the users/clients, enabling on demand executions of FaaS while maintaining the elasticity of the serverless application. Mostly CSPs plan to ensure that FaaS is tightly coupled with their platform which enables them to restrict users/clients from wandering away to other platforms/vendors. Vendor lock-in is the inability of migrating application from one CSP to another. This research focuses on the issues of limited migration flexibility and vendor lock-in within cloud-native serverless functions/FaaS, therefore it is intended to provide the implementation of Portable Function as a Service (P-FaaS) by comparing two major cloud providers and employing the Design Science Research (DSR) approach that focuses on creating practical artifacts to address real-world problems and improve understanding in the very field. It combines the rigor of scientific research with the aim of producing tangible solutions and guidelines for practical use. The research involves evaluating the services offered by the cloud providers, developing an application that abstracts vendor-specific services, and creating a migration artifact based on Large Language Model (LLM) for translating core domain functions into vendor specific format. By providing practical insights and strategies, this research aims to empower decision-makers to make informed choices, ultimately enabling smoother serverless function/FaaS migration while mitigating the risks of vendor lock-in.

Document Type

Restricted Access

Submission Type

Thesis

Download

The full text of this document is only accessible to authorized users.

Share

COinS