Making Geo-Replicated Systems Fast as Possible, Consistent when Necessary

Seminar | October 17 | 3-4 p.m. | Soda Hall, Wozniak Lounge

Speaker/Performer: Allen Clement, Max Planck Institute for Software Systems

Sponsor: Electrical Engineering and Computer Sciences (EECS)

Online services distribute and replicate state across geographically diverse data centers and direct user requests to the closest or least loaded site. While ef- fectively ensuring low latency responses, this approach is at odds with maintaining cross-site consistency. We make three contributions to address this tension. First, we propose RedBlue consistency, which enables blue operations to be fast (and eventually consistent) whi- le the remaining red operations are strongly consistent (and slow). Second, to make use of fast operation when- ever possible and only resort to strong consistency when needed, we identify conditions delineating when opera- tions can be blue and must be red. Third, we introduce a method that increases the space of potential blue op- erations by breaking them into separate generator and shadow phases. We built a coordination infrastructure called Gemini that offers RedBlue consistency, and we report on our experience modifying the TPC-W and RU- BiS benchmarks and an online social network to use Gemini. Our experimental results show that RedBlue consistency provides substantial performance gains with- out sacrificing consistency.

Biography:
Allen Clement is Faculty (tenure track) at the Max Planck Institute for Software Systems in Germany. He received his PhD from the University of Texas at Austin in 2010 under the supervision of Lorenzo Alvisi and Mike Dahlin. His research is centered around the challenges of building systems robust to the variety of things that can go wrong in computer systems: selfish or malicious behavior, faulty hardware/software/networking components, or benign concurrency. When not in the lab he is likely to be found playing Ultimate Frisbee or coaching one of the German national teams.

Event Contact: alig@cs.berkeley.edu