Practical Networked Systems for Data Centers

Speaker:        Dr. Kai CHEN
                Assistant Professor
                Department of Computer Science and Engineering
                Hong Kong University of Science and Technology

Title:          "Practical Networked Systems for Data Centers"

Date:           Monday, 28 November 2016

Time:           4:00pm - 5:00pm

Venue:          Lecture Theater F (near lifts 25/26), HKUST

Abstract:

Data centres are the main infrastructures for big data and cloud
computing. In our SING lab at HKUST, we have been working on data center
networking systems research for years. In this talk, I will start with
briefly introducing a lab-scale 100-server data center cluster we built in
the last 4 years and some representative networked system projects we
implemented based on it. Then, I will elaborate more on the project of
multi-level flow scheduling, and its adoption in practice to support
machine learning applications in a large Internet company. At last, I will
conclude my talk with the plan on connecting lab research with real-world
for more practical impacts.


******************
Biography:

Kai Chen is an Assistant Professor with CSE at HKUST, where he leads the
system networking (SING) lab. He received his PhD in Computer Science from
Northwestern University in 2012. His main research interests include
networked systems design and implementation, data center networks,
networking support for cloud and big data applications. His work has
appeared in top conferences such as SIGCOMM and NSDI, and he is actively
collaborating with large companies such as Microsoft, Tencent, and Huawei
to push his research into practice.