Brian F. Cooper

Contact information:

cooperb at google com (work)

brianfrankcooper at gmail com (personal)

I am a software engineer at Google, working on Search serving and AI infrastructure. 

Before Search, I worked on Google Workspace, focusing on storage, routing, geographical data placement, and a few other things. Before Workspace, I worked on the Spanner team. My focus was on the distributed concurrency and replication mechanism of Spanner, but I also worked on a variety of things in many parts of the system. Before Spanner, I worked on Search Quality, improving the ranking algorithm for Google web search.

I taught CS 347, Parallel and Distributed Data Management, at Stanford University in the Spring of 2015 as a visiting lecturer.

Before Google, I was a principal research scientist at Yahoo! Research doing database research. Here are some cool things I did there:

• • I was a key technical architect for PNUTS/Sherpa, Yahoo's record-oriented cloud database. The system is in production serving a variety of user-facing and back-end applications. Information about PNUTS: VLDB paper, Powerpoint overview. Our VLDB paper won the "2019 VLDB Test of Time Award" and we wrote a retrospective on the project.

  • I worked on Feeding Frenzy, a set of techniques for optimizing applications like Twitter and Facebook (and our own Yahoo! Updates) by adaptively pushing or pulling updates. Information about Feeding Frenzy: SIGMOD paper, blog post.

  • I built the Yahoo! Cloud Serving Benchmark (YCSB) for cloud serving systems like PNUTS, Cassandra, HBase and so on, and used it to draw some conclusions about the relative strengths and weaknesses of different serving systems. Information about YCSB: ACM SoCC paper, Powerpoint overview/SoCC talk, source code. Our SoCC paper won the "SoCC 10-year award" (award talk).

Before that I was an assistant professor at Georgia Tech, and before that I was a PhD student at Stanford.

In general, my interests are in massive scale software systems that handle huge amounts of data. I have worked on distributed and parallel databases, web search ranking, cloud computing, self-adaptive peer-to-peer systems, distributed streaming event processing, reliable distributed archival data storage, and XML indexing.

For more work that I have done, take a look at my Publications.

External things

In the past I've been a PC member for SIGMOD, VLDB, ICDE, ACM Middleware, and other conferences and workshops, track chair for High Performance Transaction Management for ICDE 2015, and have reviewed papers for ACM TODS, IEEE TPDS, IEEE TKDE, and other journals.

Conferences and Workshops I have helped organize:

• 2016 ACM Symposium on Cloud Computing, October 3-5, 2016, Santa Clara, California, USA (general chair)

  • First ACM Symposium on Cloud Computing (SOCC), jointly sponsored by SIGMOD and SIGOPS, June 10 & 11, 2010, Indianapolis, IN, USA (treasurer)

  • ACM/IFIP/USENIX 10th International Middleware Conference (Middleware 2009), December 2009, Illinois, USA (PC co-chair)

  • Third International Workshop on Networking Meets Databases (NetDB '07), April 10, 2007, Cambridge, MA (co-chair)

  • IEEE Workshop on Workflow and Data Flow for Scientific Applications (SciFlow 2006) (co-chair)

Links

• Increasingly outdated resume

• A short biography

• Google

• Yahoo! Research

• Georgia Tech College of Computing

• Center for Experimental Research in Computer Systems

• Stanford University Database Group

• Peer-to-peer research at Stanford

Please note: this is my personal page and does not reflect the views of Google or its management in any way.