NORMA eResearch @NCI Library

Custom Kubernetes Scheduler-based on priority scheduling for Serverless Framework

Yadav, Manali (2023) Custom Kubernetes Scheduler-based on priority scheduling for Serverless Framework. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration Manual]
Preview
PDF (Configuration Manual)
Download (744kB) | Preview

Abstract

At times, Kubernetes has gained popularity as an option for managing containerised workloads at scale. When it comes to assigning workloads to the nodes Kubernetes relies on a scheduler that takes into account a range of constraints set by workload owners and cluster administrators. While the default Kubernetes scheduler is highly configurable, it may not fully meet the requirements of serverless applications. This is because it operates in a pod scheduling mode, whereas certain serverless frameworks require co-scheduled alongside pod priority. In this research, Open Whisk is an open-source serverless framework specifically designed for a Kubernetes cluster consisting of workers deployed on the Amazon Web Service (AWS) cloud at National College of Ireland. A noteworthy contribution of this study is the development of a custom scheduler for Kubernetes aimed at enhancing pod allocation to worker nodes through an innovative scoring algorithm tailored for pods. With the proposed scheduling algorithm, the Central Processing Unit (CPU) utilisation was reduced by approx. more than 50%, and the average pod scheduling time for the custom scheduler is 0.2 sec which shows a better difference as compared to the default Kubernetes scheduler. Better performance is seen in other factors as well, like memory utilisation and throughput.

Item Type: Thesis (Masters)
Supervisors:
Name
Email
Haque, Rejwanul
UNSPECIFIED
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Cloud computing
Divisions: School of Computing > Master of Science in Cloud Computing
Depositing User: Ciara O'Brien
Date Deposited: 11 Apr 2025 13:21
Last Modified: 11 Apr 2025 13:21
URI: https://norma.ncirl.ie/id/eprint/7423

Actions (login required)

View Item View Item