Welcome!

Assistant Professor

Assistant Professor
UCLA Computer Science Department
3532E Boelter Hall
Los Angeles, CA 90095-1596
tcondie [at] cs.ucla.edu
LinkedIn Profile

My main expertise is in building large-scale distributed systems for processing massive datasets. Prior to joining UCLA, I was a Principal Scientist at Microsoft in the Cloud and Information Services Lab (CISL), and a Research Scientist at Yahoo!. Prior to my graduate studies, I was a software engineer at Oracle (Server-Technologies Division) and at Sybase. These industry experiences have led me to focus on database and distributed systems technologies.

My current research aims to unify a suite of large-scale data processing tasks on a single runtime layer, built on modern resource managers (“cloud operating systems”). The goal is to factor out commonalities in specialized systems and provide them in a single underlying runtime system, shortening the time to “market” for the next ready-to-use Big Data toolkit, which in turn will increase the availability of such tools to the broader community, impacting a broad range of scientific, engineering, national security, healthcare and business applications.


Current Research Projects

REEF: Retainable Evaluator Execution Framework

The REEF project is a collaboration with the Microsoft Cloud and Information Services Lab on building a framework for simplifying and unifying the lower layers of Big Data system stacks. REEF forms a layer above resource managers (like YARN, Mesos, Corona) and below Big Data applications (like Hadoop MapReduce, Hyracks, GraphLab, Pregel). Special consideration is given to mixed-framework, graph computations and machine learning applications.

DeML: Declarative Machine Learning

The DeML project is a collaboration with Carlo Zaniolo (UCLA) and Neoklis Polyzotis (UCSC). We explore a different approach to the development of ML tools; inspired by the principle of declarative data management. The DeML system that we are building enables the authoring and execution of ML tools in a high-level declarative language that is reminiscent to Datalog. Using DeML, a tool developer implements an ML algorithm as a declarative query over the training data. A data scientist can then invoke the query on the specific data set with any required parameters. DeML then is responsible for transparently optimizing the execution of the query over a compute platform (e.g., Amazon EC2 or SQL Azure), taking into account he characteristics of the algorithm, statistics of the data, and the available computational resources.


Selected Publications

  • V. R. Borkar, Y. Bu, M. J. Carey, J. Rosen, N. Polyzotis, T. Condie, M. Weimer, and R. Ramakrishnan, “Declarative systems for large-scale machine learning,” Ieee data eng. bull., vol. 35, iss. 2, pp. 24-32, 2012.
    [Bibtex]
    @article{DSML,
    author = {Vinayak R. Borkar and
    Yingyi Bu and
    Michael J. Carey and
    Joshua Rosen and
    Neoklis Polyzotis and
    Tyson Condie and
    Markus Weimer and
    Raghu Ramakrishnan},
    title = {Declarative Systems for Large-Scale Machine Learning},
    journal = {IEEE Data Eng. Bull.},
    volume = {35},
    number = {2},
    year = 2012,
    pages = {24-32},
    ee = {http://sites.computer.org/debull/A12june/declare.pdf},
    bibsource = {DBLP, http://dblp.uni-trier.de}
    }
  • T. Condie, N. Conway, P. Alvaro, J. M. Hellerstein, K. Elmeleegy, and R. Sears, “Mapreduce online,” in Proceedings of the 7th usenix conference on networked systems design and implementation, 2010, pp. 21-21.
    [Bibtex]
    @inproceedings{MRO,
    author = {Condie, Tyson and Conway, Neil and Alvaro, Peter and Hellerstein, Joseph M. and Elmeleegy, Khaled and Sears, Russell},
    title = {MapReduce Online},
    booktitle = {Proceedings of the 7th USENIX Conference on Networked Systems Design and Implementation},
    series = {NSDI'10},
    year = 2010,
    location = {San Jose, California},
    pages = {21--21},
    numpages = {1},
    url = {http://dl.acm.org/citation.cfm?id=1855711.1855732},
    acmid = {1855732},
    publisher = {USENIX Association}
    }
  • N. Pansare, V. R. Borkar, C. Jermaine, and T. Condie, “Online aggregation for large mapreduce jobs,” Pvldb, vol. 4, iss. 11, pp. 1135-1145, 2011.
    [Bibtex]
    @article{OAMR,
    author = {Niketan Pansare and
    Vinayak R. Borkar and
    Chris Jermaine and
    Tyson Condie},
    title = {Online Aggregation for Large MapReduce Jobs},
    journal = {PVLDB},
    volume = {4},
    number = {11},
    year = 2011,
    pages = {1135-1145},
    ee = {http://www.vldb.org/pvldb/vol4/p1135-pansare.pdf},
    bibsource = {DBLP, http://dblp.uni-trier.de}
    }
  • Y. Bu, V. R. Borkar, M. J. Carey, J. Rosen, N. Polyzotis, T. Condie, M. Weimer, and R. Ramakrishnan, “Scaling datalog for machine learning on big data,” Corr, vol. abs/1203.0160, 2012.
    [Bibtex]
    @article{ScalingML,
    author = {Yingyi Bu and
    Vinayak R. Borkar and
    Michael J. Carey and
    Joshua Rosen and
    Neoklis Polyzotis and
    Tyson Condie and
    Markus Weimer and
    Raghu Ramakrishnan},
    title = {Scaling Datalog for Machine Learning on Big Data},
    journal = {CoRR},
    volume = {abs/1203.0160},
    year = 2012,
    ee = {http://arxiv.org/abs/1203.0160},
    bibsource = {DBLP, http://dblp.uni-trier.de}
    }
  • M. Weimer, T. Condie, and R. Ramakrishnan, “Machine learning in scalops, a higher order cloud computing language,” in Nips 2011 workshop on parallel and large-scale machine learning (biglearn), Sierra Nevada, Spain, 2011.
    [Bibtex]
    @inproceedings{scalops,
    Address = {Sierra Nevada, Spain},
    Author = {Markus Weimer and Tyson Condie and Raghu Ramakrishnan},
    Booktitle = {NIPS 2011 Workshop on parallel and large-scale machine learning (BigLearn)},
    Date-Added = {2011-11-21 13:16:51 -0800},
    Date-Modified = {2011-11-21 13:26:05 -0800},
    Month = {December},
    Title = {Machine learning in ScalOps, a higher order cloud computing language},
    Url = {http://cs.markusweimer.com/2011/11/21/machine-learning-in-scalops-a-higher-order-cloud-computing-language/},
    year = 2011
    }