学术报告

当前位置: 网站首页 > 科学研究 > 学术报告 > 正文

第28期学术报告:Efficiently Inferring Top-k Elephant Flows based on Discrete Tensor Completion

时间:2019-06-25 15:52:24  作者:  点击: 107 次

28期学术报告

题  目Efficiently Inferring Top-k Elephant Flows based on

      Discrete Tensor Completion

时  间:2019年6月24日10:00

地   点8号楼425

报告人 谢鲲

博士,湖南大学信息科学与工程学院教授、博士生导师,计算机工程系副主任。香港理工大学博士后,美国纽约大学石溪分校访问学者。湖南大学岳麓学者,湖南省杰出青年基金获得者。研究方向为大数据、云计算和移动云、无线网络与移动计算。主持国家自然科学基金二项,主持湖南省自然科学基金重点基金一项。在 IEEE/ACM Transactions on Networking 》,《 IEEE Transactions on Mobile Computing 》,《 IEEE Transactions on Computers 》,《 IEEE Transactions on Wireless Communications 》,《 IEEE Transactions on Service Computing 》, INFOCOM ICDCS SECON IWQoS MASS 等发表论文 70 余篇,申请国家发明专利 27 ( 已经授权 16 ) 。荣获 2009 年布鲁塞尔世界创新科技博览会铜奖, 2006 年湖南省科学技术进步奖二等奖, 2016 年湖南省科学技术奖二等奖。

Abstract:

Finding top-k elephant flows is a critical task in network measurement, with applications such as congestion control, anomaly detection, and traffic engineering. Traditional top-k flow detection problem focuses on using a small amount of memory to measure the total number of packets or bytes of each flow. Instead, we study a challenging problem of inferring the top-k elephant flows in a practical system with incomplete measurement data as a result of sub-sampling for scalability or data missing. The recent study shows it is promising to more accurately interpolate the missing data with a 3-D tensor compared to that based on a 2-D matrix. Taking full advantage of the multilinear structures, we apply tensor completion to first recover the missing data and then find the top-k elephant flows. To reduce the computational overhead, we propose a novel discrete tensor completion model which uses binary codes to represent the factor matrices. Based on the model, we further propose three novel techniques to speed up the whole top-k flow inference process: a discrete optimization algorithm to train the binary factor matrices, bit operations to facilitate quick missing data inference, and simplifying the finding of top-k elephant flows with binary code partition. In our discrete tensor completion model, only one bit is needed to represent the entry in the factor matrices instead of a real value (32 bits) needed in traditional tensor completion model, thus the storage cost is reduced significantly. Extensive experiments using two real traces demonstrate that compared with the state of art tensor completion algorithms, our discrete tensor completion algorithm can achieve similar data inference accuracy using significantly smaller time and storage space.

 

地址:嘉兴市嘉杭路118号 邮编:314001 E-Mail:jx83640102@163.com
版权所有 Copyright(C)2011 嘉兴学院数理与信息工程学院