雲端運算中能源效率及綠色能源的研究

Abstract

One of the key questions that should be asked when making technological innovation is whether or not one should make the innovation. With the breakout success of cloud computing and mobility computing devices such as smart phones, tablets, and laptops the question is more of how quickly innovation can get to the market rather than should it get to the market. Due to the inherent limitations of mobile devices, a massive cloud based infrastructure is needed behind the scenes to enable ubiquity of apps and connectivity. The cloud application and mobility fields are continuously growing as is the cloud infrastructure powering and keeping the lion’s share of the data. Here, the question not being asked is what is the real impact on the environment from cloud computing and how can this be mitigated. In a hope to make more informed decisions on the planning and commissioning of cloud computing data centers and data centers, this thesis explores the cloud computing landscape, energy efficiency, climate change and the intersection of each. There are great opportunities to reduce power consumption in cloud computing to save energy. There are also great opportunities to strategically apply green energy to the cloud. This thesis proposes a weighted levelized cost of energy metric to quantify the real cost and impact of different power sources. This is done at the global level as well as the local level in the United States and China.

With a weighted levelized cost of energy, this thesis then explores regional opportunities within the United States and considers power mix migration scenarios to improve overall weighted cost. The final step is to create a data center model and investigate permutations of power mix and location in order to better understand cost impact savings opportunities.