|
" Towards Reliability Evaluation and Integration in Cloud Resource Management "
Alam, A. B. M. Bodrul
Zulkernine, Mohammad
| Document Type
|
:
|
Latin Dissertation
|
| Language of Document
|
:
|
English
|
| Record Number
|
:
|
1059094
|
| Doc. No
|
:
|
TL58211
|
| Main Entry
|
:
|
Alam, A. B. M. Bodrul
|
| Title & Author
|
:
|
Towards Reliability Evaluation and Integration in Cloud Resource Management\ Alam, A. B. M. BodrulZulkernine, Mohammad
|
| College
|
:
|
Queen's University (Canada)
|
| Date
|
:
|
2020
|
| Degree
|
:
|
Ph.D.
|
| student score
|
:
|
2020
|
| Note
|
:
|
136 p.
|
| Abstract
|
:
|
The adoption of cloud computing technology is increasing day by day due to its enormous services and low cost. Customers transfer their businesses to cloud because of the beneficial cloud features such as scalability, high performance, on-demand, and pay-per-use service model. However, transferring services to cloud adds a new level of risk due to loss of control, which makes service reliability an essential driving factor of the cloud market today, especially in light of the recent cloud failures and outages that raise customers' concerns. In this thesis, the goal is to propose a reliability evaluation model and integrate the developed model in cloud resource management such as Virtual Machine (VM) allocation, VM migration, and cloud federation formation in order to increase the cloud service reliability. In this thesis, firstly a cloud reliability evaluation model is proposed. Some types of failures from different domains of the cloud environment are considered to evaluate cloud reliability. To propose the evaluation model, a classification strategy for cloud failures is also outlined. Secondly, to show the impact of the integration of the proposed model, a multi-objective placement model for interdependent VMs in cloud is proposed while considering both reliability and some quality of service. A multi-objective genetic algorithm is used to solve the placement problem heuristically. Thirdly, a Markov-based failure prediction model is proposed to anticipate the failures of cloud servers. The proposed prediction model is then integrated into a VM migration model in a multi-cloud setting to maximize cloud reliability while reducing VM communication delay. The VM migration problem is solved optimally and heuristically using the Artificial Bee Colony (ABC) algorithm. Finally, a reliability-based cloud federation model is proposed using a hedonic coalition formation game based on a reliability-driven utility function. All proposed models will serve as a guide to both customers and cloud service providers towards the achievement of reliable resource allocation, migration, and federation formation. The effectiveness of all the models has been demonstrated through experiments. The evaluation shows that the models are computationally efficient and achieve higher cloud reliability.
|
| Descriptor
|
:
|
Algorithms
|
|
|
:
|
Classification
|
|
|
:
|
Cloud computing
|
|
|
:
|
Communication
|
|
|
:
|
Customer services
|
|
|
:
|
Failure analysis
|
|
|
:
|
Fault tolerance
|
|
|
:
|
Optimization
|
|
|
:
|
Software
|
|
|
:
|
Workloads
|
| Added Entry
|
:
|
Zulkernine, Mohammad
|
| Added Entry
|
:
|
Queen's University (Canada)
|
| |