Energy and carbon-efficient resource management in geographically distributed cloud data centers
AffiliationComputing and Information Systems
MetadataShow full item record
Document TypePhD thesis
Access StatusOpen Access
© 2017 Dr Atefeh Khosravi
Cloud computing provides on-demand access to computing resources for users across the world. It offers services on a pay-as-you-go model through data center sites that are scattered across diverse geographies. However, cloud data centers consume huge amount of electricity and leave high amount of carbon footprint in the ecosystem. This makes data centers responsible for 2% of the global CO2 emission, the same as the aviation industry. Therefore, having energy and carbon-efficient techniques for distributed cloud data centers is inevitable. Cloud providers while efficiently allocating computing resources to users, should also meet their required quality of service. The main objective of this thesis is to address the problem of energy and carbon efficient resource management in geographically distributed cloud data centers. It focuses on the techniques for VM placement, investigates the parameters with largest effect on the energy and carbon cost, migration of VMs between data center sites to harvest renewable energy sources, and prediction of renewable energy to maximize its usage. The key contributions of this thesis are as follows: (1) A VM placement algorithm to optimally select the data center and server to reduce energy consumption and carbon footprint with considering energy and carbon related parameters. (2) A dynamic method for the initial placement of VMs in geographically distributed cloud data centers that simultaneously considers energy and carbon cost and maximizes renewable energy utilization at each data center to minimize the total cost. (3) Variations of VM placement methods, which explore the effects of different parameters in minimizing energy and carbon cost for a cloud computing environment. (4) The optimal offline algorithm and two online algorithms, which exploit available renewable energy levels across distributed data center sites for VM migration to minimize total energy cost and maximize renewable energy usage. (5) A prediction model for renewable energy availability at data center sites to incorporate into online VM migration algorithm and maximize renewable energy usage.
Keywordscloud computing; data center; green computing; renewable energy; VM placement; VM migration; carbon footprint
- Click on "Export Reference in RIS Format" and choose "open with... Endnote".
- Click on "Export Reference in RIS Format". Login to Refworks, go to References => Import References