Service selection, comparison, and recommendation with continuous quality of service assurance and optimization in cloud computing environment
Miranda Zhang (The Australian National University)
COMPUTER SYSTEMS SEMINAR These Proposal Review SeminarDATE: 2013-08-23
TIME: 12:00:00 - 12:30:00
LOCATION: CSIT N101
CONTACT: JavaScript must be enabled to display this email address.
ABSTRACT:
Cloud computing has become a very important part of the knowledge economy and is certain to become one of the biggest drivers of economic growth over the next decade. Gartner forecasts public cloud services to be worth $109 billion this year, while the EU expects the cloud to add as much as E160 billion ($206 billion) to annual GDP between now and 2020.1
The success of Cloud computing has encouraged many application developers to consider migrating their applications to the Cloud.2 Any such migration of applications (e.g. multi-layered enterprise application, scientific experiments, video-on-demand streaming application, etc.) to the Cloud demands selecting the best mix of services from an abundance of possibilities. Cloud service selection decision has to cater for conflicting criteria, e.g. maximise QoS while minimising cost. The problem is further aggravated by the lack of standard and uniform representation of IaaS. In order to address these aforementioned problems, my thesis will explore semi-automated, extensible, and ontology-based approach to infrastructure service discovery and selection. In our preliminary work, we developed a model that make the best attempt to capture the important characteristics of IaaS, and allow user to define multiple quantitative and qualitative requirements which are then matched against our knowledge base to compute possible best fit combinations, solutions are ordered by ranking which are calculated using AHP.
In the future, we would like to provide smarter decision support by including SLA, legal compliance and software PaaS features into consideration. We are also improving the data gathering and updating mechanism.
Furthermore, the system we are proposing may also benefit the Cloud provider, by providing analyse of the market and demand, our system can potentially recommend what price the providers can set their service to.
Aiming to eliminate potential bottlenecks that limit the ability of general users to take advantage of the cloud computing future, I hope my research will drive even greater adoption of the cloud and boost the expansion of the cloud economy.
BIO:
Miranda Zhang is a PhD student in the Computer System Group at RSCS. She holds a Bachelor of Software Engineering degree from TheAUniversity of New South Wales, Sydney. She had done a half year internship with CSIRO in 2012 before start her PhD.
