Tyson Condie

 

I've been a computer science PhD student at Berkeley since Fall 2004. My research focuses on large scale

data management systems. I am particularly interested in the query optimization and work scheduling aspects

of these systems. 


My advisor is Prof. Joe Hellerstein

I currently collaborate with research scientists at Y! Labs on Hadoop related projects.

 

Publications 

 

·      MapReduce Online.
  Authors: Tyson Condie, Neil Conway, Peter Alvaro, Joseph M. Hellerstein, Khaled Elmeleegy, and Russell Sears.
  UC Berkeley Technical Report No. UCB/EECS-2009-136, 2009. In submission.

 

·      BOOM: Data-Centric Programming in the Datacenter.
  Authors: Peter Alvaro, Tyson Condie, Neil Conway, Khaled Elmeleegy, Joseph M. Hellerstein, and Russell Sears.
  UC Berkeley Technical Report No. UCB/EECS-2009-113, 2009. In submission.

 

·      I Do Declare: Consensus in a Logic Language.
  Authors: Peter Alvaro, Tyson Condie, Neil Conway, Joseph M. Hellerstein, and Russell Sears.
  In Proceedings of the SOSP Workshop on Networking Meets Databases (NetDB), 2009.

 

·      Evita Raced: Metacompilation for Declarative Networks.
  Authors: Tyson Condie, David Chu, Joseph M. Hellerstein, and Petros Maniatis.
  In 34th International Conference on Very Large Data Bases (VLDB), 2008.

 

·      Public Health for the Internet (PHI).
  Authors: Joseph M. Hellerstein, Tyson Condie, Minos Garofalakis, Boon Thau Loo, Petros Maniatis,

  Timothy Roscoe, and Nina Taft.

  In CIDR, 2007.

 

·      ROFL: Routing On Flat Labels.
  Authors: Matthew Caesar, Tyson Condie, Jayanthkumar Kannan, Karthik Lakshminarayanan,

  Ion Stoica, Scott Shenker.
              In ACM SIGCOMM International Conference on Computer Communication, 2006.

 

·      Declarative Networking: Language, Execution and Optimization.
  Authors: Boon Thau Loo, Tyson Condie, Minos Garofalakis, David A. Gay, Joseph M. Hellerstein, Petros Maniatis,

  Raghu Ramakrishnan, Timothy Roscoe and Ion Stoica.

  In ACM-SIGMOD International Conference on Management of Data, 2006.

 

·      Induced Churn as Shelter from Routing-Table Poisoning.
  Authors: Tyson Condie, Varun Kacholia, Sriram Sankararaman, Joseph M. Hellerstein and Petros Maniatis.
  In Network and Distributed System Security (NDSS), 2006.

 

·      Implementing Declarative Overlays.
  Authors: Boon Thau Loo, Tyson Condie, Joseph M. Hellerstein, Petros Maniatis, Timothy Roscoe, and Ion Stoica.
  In 20th ACM Symposium on Operating Systems Principles (SOSP), 2005.

 

·      Finally, a use for componentized transport protocols.
  Authors: Tyson Condie, Joseph M. Hellerstein, Petros Maniatis, Sean Rhea, and Timothy Roscoe.
  In HotNets IV. 2005.

 

·      LSH Forest: self-tuning indexes for similarity search.
  Authors: Mayank Bawa, Tyson Condie, and Prasanna Ganesan.
  In 14th International Conference on World Wide Web (WWW), 2005.

 

·      Non-Cooperation in Competitive P2P Networks.
  Authors: Beverly Yang, Tyson Condie, Sepandar Kamvar and Hector Garcia-Molina.
  In Distributed Computing Systems, 2005.

 

·      Adaptive Peer-To-Peer Topologies.
  Authors: Tyson Condie, Sepandar Kamvar and Hector Garcia-Molina.
  In International Conference on Peer-to-Peer Computing, 2004.

 

·      Simulating a File-Sharing P2P Network.
  Authors: Mario Schlosser, Tyson Condie, and Sepandar Kamvar.
  In Workshop on Semantics of P2P and Grid Computing, 2002.

 

Research Projects

 

BOOM Berkeley Orders Of Magnitude

BOOM is an exploration of how data-centric programming methodology can

make Cloud Computing an accessible platform for developer innovation. I

am involved with all aspects of this project.

 

Pig

Pig is an infrastructure to support ad-hoc analysis of very large data sets. I focused

on the development of an optimization framework for this project.

 

P2

P2 is a system that uses a high-level declarative language to express overlay

networks in a highly compact and reusable form. P2 is part of a more general effort

to revisit networking technology through the lens of database query processing. I

am involved with all aspects of this project.

 

Titanium

Titanium is an explicitly parallel dialect of Java developed at UC Berkeley to support

high-performance scientific computing on large-scale multiprocessors, including massively

parallel supercomputers and distributed-memory clusters with one or more processors per

node. I focused on the development of distributed locking protocols for this project.

 

 

Contact

 

tcondie at cs.berkeley.edu

Resume

419 Soda Hall,
Berkeley, CA 94720-1776