NORMA eResearch @NCI Library

Dynamic Resources allocation using Priority Aware scheduling in Kubernetes

Shelar, Prasad Lahu (2019) Dynamic Resources allocation using Priority Aware scheduling in Kubernetes. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (3MB) | Preview

Abstract

In current IT infrastructure, Microservices based architecture provides the loosely coupled services for the development, deployment of the application. In a microservice architecture container-based application runs on the cloud service provider to achieve business continuity. Hundreds of container-based applications are deployed on a daily basis in the service provider. In order to manage the containers and also to scale the application, there is a need for container orchestrator and currently the leading orchestrator tools in the market are Kubernetes, Dockerswarm and ApacheMesos. By default, Kubernetes schedule containers use the Bin-packing algorithm. The scheduler allocates the resources as per the availability vs demand in First fit first manner. In this research, we are trying to explore Application-aware scheduling. The mechanism involves allocating the containers based on priority with the custom python-based scheduler in Kubernetes architecture. Our results show that priority aware scheduler can allocate the resources without affecting the services as well as Kubernetes default scheduler.

Item Type: Thesis (Masters)
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: CAOIMHE NI MHAICIN
Date Deposited: 20 Mar 2020 14:56
Last Modified: 20 Mar 2020 14:56
URI: http://trap.ncirl.ie/id/eprint/4137

Actions (login required)

View Item View Item