alt text 

Stanford University Networking Seminar


Yuliang Li (University of Southern California)
Fast and efficient flow and loss measurement for data center networks

12:15pm, Monday, May 23 2016
Packard 202

About the Talk

Data center networks need to monitor all flow's information on a fine time scale for debugging and profiling such as transient loops/blackholes detection, packet loss analysis, attack detection, etc. However, existing tools like NetFlow falls short because it requires sampling to reduce its high overhead. We design FlowRadar, a new way to maintain flows and their counters at a fine time scale, which scales to a large number of flows with small memory and bandwidth overhead. The key idea of FlowRadar is to encode per-flow counters with a small memory and constant insertion time at switches, and then to leverage the computing power at the remote collector to perform network-wide decoding and analysis of the flow counters. In addition to providing flow information, we are also working on a new tool LossRadar that can detect each individual loss with the time and locations information in near real time.

About the Speaker

Yuliang Li is a 2nd year Ph.D. student in the computer science department at the University of Southern California. He received his B.E. in computer science and technology from Tsinghua University in 2014. He has interned at Microsoft Azure. His research interest is in data center networking and software-defined networking.