Massive Graphs in Clusters

Many of today's data-intensive application domains, including searches on social networks like Facebook and protein matching in bioinformatics, require us to answer complex queries on highly-connected data. This data is often represented as a graph of data objects densely connected by edges. The UCSB Massive Graphs in Clusters (MAGIC) project is focused on developing software infrastructure that can efficiently answer queries on extremely large graph datasets. The MAGIC software will provide an easy to use interface for searching and analyzing data, and manage the processing of these queries to efficiently take advantage of computing resources like large data centers.

Advances from this project will include techniques to distribute and manage data across large datacenters, high-level interface for writing graph queries, and a software infrastructure that builds on and extends cluster-based software systems such as MapReduce and Dryad.

The MAGIC project is funded by the National Science Foundation under the CLuE (Cluster Exploratory) program.

Recent News

  • April 30, 2010: Dr. Alessandra Sala presented her paper at WWW 2010 on high-fidelity graph models for measurement- calibrated social networking experiments. Paper is available on the publications page.
  • March 2, 2010: Dr. Xifeng Yan and Sudipto Das presented their papers at ICDE 2010, one on Top-K aggregation queries and one on anonymizing weighted graphs. Papers are available on the publications page.
  • February 22, 2010: Krishna Puttaswamy opened up the ACM HotMobile 2010 workshop by presenting his work on privacy protection for location-based social network applications.
  • December 3, 2009: Krishna Puttaswamy presented his work on protecting privacy of content-sharing social applications at ACM CoNEXT.
  • April 3, 2009: Christo Wilson presented his work on Facebook user interactions and implications on social applications.
  • March 28, 2009: Magic website goes live!