Streaming Algorithms for High Throughput Massive Datasets

Streaming Algorithms;Randomized Algorithms;Big Data;Computer Science

Vorsanger, Gregory AlexanderBraverman, Vladimir ;

Johns Hopkins University

关键词: Streaming Algorithms;    Randomized Algorithms;    Big Data;    Computer Science;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/40831/VORSANGER-DISSERTATION-2017.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

The field of streaming algorithms has enjoyed a deal of focus from the theoretical computer science community over the last 20 years. Many great algorithms and mathematical results have been developed in this time, allowing for a broad class of functions to be computed and problems to be solved in the streaming model. In the same amount of time, the amount of data being generated by practical computer systems is simply staggering. In this thesis, we focus on solving problems in the streaming model that have a unified goal of being relevant to practical problems outside of the theory community. In terms of a common technical thread throughout this work, the theme here is an attention to runtime and the ability to handle large datasets that not only challenge in terms of memory available, but also in the throughput of the data and the speed at which the data must be processed.We provide these solutions in the form of both theoretical algorithm and practical systems, and demonstrate that using practice to drive theory, and vice versa, can generate powerful new approaches for difficult problems in the streaming model.