构建高性能集群框架:从大数据分析到深度学习系统

2019.05.24

投稿:周时强部门:计算机工程与科学学院浏览次数:

活动信息

时间: 2019年05月24日 15:30

地点: 校本部东区计算机大楼1001室

报 告 人:王堃,加州大学洛杉机分校

报告时间:05月24日(周五)15:30~16:30

报告地点:校本部东区计算机大楼1001室

邀 请 人:李卫民 副教授


报告摘要: Data-intensive applications often involve the multi-level management in data center networks, where distributed operating systems for parallel computing are deployed. Nowadays, production requirements for operating modern applications usually resort to the assistance of the underlay infrastructure we called cluster frameworks. As a result, building a high-performance cluster framework for specific applications has become one of the most crucial issues in the relevant research fields. In this talk, we intend to give a comprehensive overview about cluster framework to audiences who are interested in this regard. Two popular application categories, traditional big data analytics and modern deep learning technologies, will be presented through the detailed analysis of algorithm design and prototype implementation. Additionally, performance evaluation of both trace-driven simulation and realistic deployment with comparison of some state-of-the-art counterparts will be illustrated. This talk will help the audiences quickly understand the research progress and fundamental basis of cluster frameworks, big data analytics, deep learning and other topics concerned.. 


报告人简介:

Dr. Kun Wang received two Ph.D. degrees from Nanjing University of Posts and Telecommunications, China in 2009 and from the University of Aizu, Japan in 2018, respectively, both in Computer Science. He was a Postdoc Fellow in UCLA, USA from 2013 to 2015, and a Research Fellow in the Hong Kong Polytechnic University, Hong Kong, from 2017 to 2018. He is currently a Senior Research Professor in UCLA. His research interests are mainly in the areas of big data, datacenter, blockchain, and distributed systems with over 100 papers published in major conferences and journals, including IEEE TIP, TC, TPDS, ToN, TMC and ACM Mobicom, ACM Mobisys, IEEE ICDCS, IEEE IPDPS, as well as 12 ESI High Cited Papers. He is the recipient of four Best Paper Awards at IEEE GLOBECOM 2016,IEEE TCGCC 2018, IEEE TCBD 2019, and IEEE ISJ (IEEE Systems Journal) 2019. He serves as Associate Editor of IEEE Access, Editor of Journal of Network and Computer Applications, and Guest Editors of IEEE Network, IEEE Access, Future Generation Computer Systems, Peer-to-Peer Networking and Applications, Journal of Internet Technology, and Future Interne。