Yahoo! at VLDB 2008

The database community congregated in beautiful Auckland, New Zealand for the 34th International Conference on Very Large Databases (VLDB 2008), held August 24 to 28. Yahoo! walked away with a total of 10 accepted papers, and several Yahoo! researchers served on various program committees.

The conference brought together members of the database research community, including representatives from Yahoo!, Google, Microsoft, Oracle, SAP, IBM, Hewlett-Packard, and various academic organizations from around the world. This year’s event saw a strong interest in industrial efforts in general, and an understanding that there is enormous potential for research on problems faced by Yahoo!, Google, Amazon and other large Internet companies. There was also a focus on query optimization, privacy, indexing and performance.

Researcher Jayavel Shanmugasundaram had 3 research track papers that were presented at the conference. Researcher Brian Cooper chaired an industrial session and a research session. He also presented PNUTS – a current database platform project at Yahoo! Research. The talk generated some buzz on cloud computing and conversation about the general debate in the database community between generic databases (one size fits all), and specialized databases that are optimized for specific needs.

There was also a lot of student interest in getting access to a “cloud” at Yahoo! to do research, and a lot of curiosity about the Yahoo! cluster partnership with CMU. Conference attendees seemed to understand that the future of Web applications lies in building them on the cloud.

Mark D. Hill of the University of Wisconsin - Madison gave a keynote on the recent explosion of transactional memory and its goals and implementation options. The other keynote talk was given by Justin Zobel of the University of Melbourne. In his talk, he linked computer science and the quality of life by explaining that innovations in databases have enormous potential for impact on health and biomedical research because medical research is generating large volumes of data that cannot be processed with existing algorithms.

Downtown Auckland provided a vibrant metropolitan backdrop for the conference -- a fitting reflection of a vibrant and active database community.

Yahoo! Accepted Papers:

Efficient Network-Aware Search in Collaborative Tagging Sites
Michael Benedikt, Sihem Amer Yahia, Laks Lakshmanan, Julia Stoyanovich

Efficient Top-K Processing over Query-Dependent Functions
Lin Guo, Sihem Amer Yahia, Raghu Ramakrishnan, Jayavel Shanmugasundaram, Utkarsh Srivastava, Erik Vee
BayeStore: Managing Large, Uncertain Data Repositories with

Probabilistic Graphical Models
Daisy Zhe Wang, Eirinaios Michelakis, Minos Garofalakis, Joseph Hellerstein

Scheduling Shared Scans of Large Data Files
Parag Agrawal, Daniel Kifer, Christopher Olston
Simrank++: Query Rewriting through Link Analysis of the

Click Graph
Ioannis Antonellis, Hector Garcia-Molina, Chi-Chao Chang

WYSIWYG Development of Data Driven Web Applications
Fan Yang, Chavdar Botev, Nitin Gupta, Elizabeth Churchill, George Levchenko, Jayavel Shanmugasundaram

Scalable Ranked Publish/Subscribe
Ashwin Machanavajjhala, Erik Vee, Minos Garofalakis, Jayavel Shanmugasundaram

Scalable Query Result Caching for Web Applications
Charles Garrod, Amit Manjhi, Bruce Maggs, Todd Mowry, Anthony Tomasic, Christopher Olston, Anastasia Ailamaki

Relaxation in Text Search using Taxonomies
Marcus Fontoura, Vanja Josifovski, Ravi Kumar, Christopher Olston, Sergei Vassilvitskii, Andrew Tomkins

PNUTS: Yahoo!'s hosted data serving platform
Brian Cooper, Raghu Ramakrishnan, Utkarsh Srivastava, Adam Silberstein, Phil Bohannon, Hans-Arno Jacobsen, Nick Puz, Daniel Weaver, Ramana Yerneni