dc.contributor.advisor |
O'Sullivan, M |
en |
dc.contributor.advisor |
Walker, C |
en |
dc.contributor.author |
Hamling, Isaac |
en |
dc.date.accessioned |
2019-03-21T01:25:25Z |
en |
dc.date.issued |
2019 |
en |
dc.identifier.uri |
http://hdl.handle.net/2292/46232 |
en |
dc.description.abstract |
The idea to use cloud computing and virtualisation to supply video games to end users known as cloud gaming has been growing in popularity. Internet cafes are one potential application for cloud gaming. Internet cafes are a large market in China which has over 185,000 internet cafes each with an average of 120 seats. These internet cafes have significant resource inefficiency which can be improved using a cloud gaming model to supply internet cafe users. A system is presented for a cloud-based internet cafe and algorithms for maximising resource utilisation in cloud gaming. The cloud-based internet cafe model replaces the traditional desktop computers in an internet cafe with servers, thin clients, and virtual machines with specifications designed to meet specific user demands. Virtual machines are run on the servers to supply internet cafe users with their desired game or service. Important decisions need to be made: which users to accept, and upon which servers they should be placed. This problem is the cloud gaming resource allocation problem. An integer programming model is formulated for solving the offline cloud gaming resource allocation problem for cloud-based internet cafes. This offline model shows that moving to a cloud-based internet cafe improves daily profits over the current zoned internet cafe model utilised in China and significantly improves resource utilisation. A further three algorithms are presented to solve real world demand in a cloud-based internet cafe. A prebooking system for solving a semi-online version of the problem. This algorithm is used to place users who book seats in the cloud-based internet cafe in advance. Online greedy and competitive algorithms are also presented for solving the online resource allocation problem for cloud-based internet cafes. All three algorithms show competitive performance when compared to the offline optimal allocations with exact performance depending the user demand profile. The cloud-based internet cafe shows the ability to improve profits, and resource efficiency in internet cafes. The four algorithms developed can be applied to different types of internet cafes. The competitive algorithm performs best in busy internet cafes, while the greedy algorithms are better when they are less busy. The prebooking system is most useful for an internet cafe with regular customers. While the offline integer program can be used for future planning or planned events. All four algorithms and cloud gaming hardware can combine to form a cloud-based internet cafe system for building and operating a cloud-based internet cafe. |
en |
dc.publisher |
ResearchSpace@Auckland |
en |
dc.relation.ispartof |
PhD Thesis - University of Auckland |
en |
dc.relation.isreferencedby |
UoA99265139612202091 |
en |
dc.rights |
Items in ResearchSpace are protected by copyright, with all rights reserved, unless otherwise indicated. Previously published items are made available in accordance with the copyright policy of the publisher. |
en |
dc.rights.uri |
https://researchspace.auckland.ac.nz/docs/uoa-docs/rights.htm |
en |
dc.rights.uri |
http://creativecommons.org/licenses/by-nc-sa/3.0/nz/ |
en |
dc.title |
Optimal Orchestration of a Cloud-Based Internet Cafe |
en |
dc.type |
Thesis |
en |
thesis.degree.discipline |
Operations Research |
en |
thesis.degree.grantor |
The University of Auckland |
en |
thesis.degree.level |
Doctoral |
en |
thesis.degree.name |
PhD |
en |
dc.rights.holder |
Copyright: The author |
en |
dc.rights.accessrights |
http://purl.org/eprint/accessRights/OpenAccess |
en |
pubs.elements-id |
766525 |
en |
pubs.record-created-at-source-date |
2019-03-21 |
en |
dc.identifier.wikidata |
Q112948686 |
|