NORMA eResearch @NCI Library

A Novel Approach for Fair Resource Allocation in Kubernetes

-, Anjalee (2020) A Novel Approach for Fair Resource Allocation in Kubernetes. Masters thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[img]
Preview
PDF (Configuration manual)
Download (1MB) | Preview

Abstract

The computing systems and networks commonly uses resource allocation standards and protocols to secure efficiency and fairness when allocating resources among different users. A problem arrives while selecting allocation rules when resources are limited and not enough to fully meet the demand on its entirety. It is significant to consider their QoS in Kubernetes when designing resource allocation rules. By fairness, it means that all users obtain impartial amount of system resources. Instead of providing fairness to users with good channel conditions, a trade-off is required between maximising the fairness between users and maximising the system throughput. Previous works in the literature of resource allocation algorithms overlook the fairness criteria and enables users possessing good channel conditions to acquire most of the resources. In this research CPU and Memory requests and limits are configured at the time of creation in case of Kubernetes default scheduler. When using Fully Fair Multi-Resource Allocation (FFMRA) algorithm, the resource limits will be applied after the scheduling for better resource allocation. Using this custom scheduler improves the resource utilization when two web applications are considered as proved in the results. The Grafana is used for monitoring the nodes and pods activities.

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: Dan English
Date Deposited: 28 Jan 2021 13:09
Last Modified: 28 Jan 2021 13:09
URI: http://norma.ncirl.ie/id/eprint/4527

Actions (login required)

View Item View Item