Physical Sciences and Mathematics Commons^™

Towards Optimal Resource Provisioning For Running Mapreduce Programs In Public Clouds, Fengguang Tian

Browse all Theses and Dissertations

Running MapReduce programs in the public cloud introduces the important problem: how to optimize resource provisioning to minimize the financial charge for a specific job? In this thesis, We study the whole process of MapReduce processing and build up a cost function that explicitly models the relationship between the amount of input data, the available system resources (Map and Reduce slots), and the complexity of the Reduce function for the target MapReduce job. The model parameters can be learned from test runs with a small number of nodes on a small amount of data. Based on this cost model, we …

Parallel Sorting On Multi-Core Architecture, Wei Wang

Browse all Theses and Dissertations

With the limitations given by the power consumption (power wall), memory wall and the instruction level parallelism, the computing industry has turned its direction to multi-core architectures. Nowadays, the multi-core and many-core architectures are becoming the trend of the processor design. But how to exploit these architectures is the primary challenge for the research community. To take advantage of the multi-core architectures, the software design has undergone fundamental changes.

Sorting is a fundamental, important problem in computer science. It is utilized in many applications such as databases and search engines. In this thesis, we will investigate and auto-tune two parallel …