J
Joseph M. Hellerstein
Researcher at University of California, Berkeley
Publications - 313
Citations - 39413
Joseph M. Hellerstein is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Query optimization & Query language. The author has an hindex of 91, co-authored 300 publications receiving 37800 citations. Previous affiliations of Joseph M. Hellerstein include Carnegie Mellon University & IBM.
Papers
More filters
Posted Content
The Asilomar Report on Database Research
Phil Bernstein,Michael L. Brodie,Stefano Ceri,David J. DeWitt,Michael J. Franklin,Hector Garcia-Molina,Jim Gray,Jerry Held,Joseph M. Hellerstein,H. V. Jagadish,Michael Lesk,Dave Maier,Jeffrey F. Naughton,Hamid Pirahesh,Michael Stonebraker,Jeffrey D. Ullman +15 more
TL;DR: The database research community should embrace a broader research agenda — broadening the definition of database management to embrace all the content of the Web and other online data stores, and rethinking the authors' fundamental assumptions in light of technology shifts.
Proceedings Article
The Architecture of PIER: an Internet-Scale Query Processor
Journal Article
Adaptive Query Processing: Technology in Evolution.
Joseph M. Hellerstein,Michael J. Franklin,Sirish Chandrasekaran,Amol Deshpande,Kris Hildrum,Samuel Madden,Vijayshankar Raman,Mehul A. Shah +7 more
TL;DR: A survey of prior work on adaptive query processing is presented, focusing on three characterizations of adaptivity: the frequency of adaptability, the effects of Adaptivity, and the extent of adaptiveness, to set the stage for research in the Telegraph project.
Proceedings ArticleDOI
Behavior of database production rules: termination, confluence, and observable determinism
TL;DR: The Starburst Rule System as discussed by the authors provides static analysis methods for determining whether a set of database production rules are (1) guaranteed to terminate; (2) guaranteed of producing a unique final database state; and (3) guaranteeing to produce a unique stream of observable actions; when the analysis determines that one of these properties is not guaranteed, it isolates the rules responsible for the problem and determines criteria that, if satisfied, guarantee the property.
Proceedings ArticleDOI
Highly available, fault-tolerant, parallel dataflows
TL;DR: A technique that masks failures in a cluster to provide high availability and fault-tolerance for long-running, parallelized dataflows that can use these dataflows to implement a variety of continuous query applications that require high-throughput, 24x7 operation.