| As cloud computing keeps developing, more and more Data Centers (DCs) need to be deployed in order to satisfy growing service demands, which result in serious challenges on energy sustainability in terms of CO2 emissions caused by large amount of non-renewable energy consumption inside DCs. This situation can be relieved by placing DCs near renewable energy sources such as wind farms, solar panels, etc. However, renewable-energy-abundant places are often geographically unsuitable for DCs because some of them are located at the edge of the network topology and far from traffic sources. As a result, in order to satisfy traffic demand and take advantage of renewable energy at the same time, more DCs are needed. But a large number of DCs will increase Capital Expenditure (CAPEX) and Operational Expenditure (OPEX), which is not preferred by InPs (Infrastructure Providers).;In this thesis, we propose a method of solving DC placement problem in virtual-machine-migration-enabled optical cloud networks with the aim of jointly minimizing non-renewable DC energy consumption and the number of DCs. Lexicographic optimization is utilized to balance the two objectives. |
