Javascript must be enabled for the correct page display

An improved model for the allocation of carbon emissions among the tenants of cloud services

Dijkstra, Albert (2023) An improved model for the allocation of carbon emissions among the tenants of cloud services. Master's Thesis / Essay, Computing Science.

[img]
Preview
Text
An improved model for the allocation of carbon2.pdf

Download (2MB) | Preview
[img] Text
toestemming.pdf
Restricted to Registered users only

Download (135kB)

Abstract

A lot of current software services use technologies facilitated by cloud computing. Cloud computing produces an increasing amount of emissions. Estimating the carbon footprint of cloud computing can be done by various models. However, it remains unclear which stakeholders bear responsibility for which part of the emissions, leaving the question of accountability unanswered. In cooperation with BT Global Services, this project continued the development of a model to estimate the carbon emissions of cloud-based software services, and distribute them fairly among stakeholders. The estimate is made by splitting energy consumption into static energy usage, which cloud resources consume while idling, and dynamic energy usage, which is based on the actual usage pattern. Based on this, a lower and upper bound, together with a set of policies for deciding these bounds is defined with respect to which part of the energy consumption the stakeholder is responsible for. This provides the involved stakeholders with more insight on their footprint. On top of this, a new metric which clarifies how energy efficient a tenant uses a service in relation to other tenants is proposed. The use of this metric was evaluated by different stakeholders with respect to its efficacy.

Item Type: Thesis (Master's Thesis / Essay)
Supervisor name: Andrikopoulos, V. and Feitosa, D.
Degree programme: Computing Science
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 26 Jul 2023 06:22
Last Modified: 26 Jul 2023 06:22
URI: https://fse.studenttheses.ub.rug.nl/id/eprint/30903

Actions (login required)

View Item View Item