Research Interests and Projects for Philip Bohannon

(return to home page)

CARBON/Snapper

(Data Exchange and Cleaning.) We are looking at a variety of issues in data transformation, XML publishing, schema matching, and constraint-based data cleaning. This work is motivated by a variety of problems faced in the delivery of inventory and optimization services to network providers.
With WenfeiFan, Michael Flaster, P.P.S. Narayan and Rajeev Rastogi .

Papers:

ROLEX

(Relational On-Line Exchange with XML.) The goal of this project is to allow new web and eCommerce applications to interoperate with traditional relational business applications through consistent, simultaneous access to XML and relational views of the same data. ROLEX capitalizes on main-memory database technology ( DataBlitz) to provide extremely high-speed access, obviating the need for application-level caches of database data.
With Hank Korth and P.P.S. Narayan .

Papers:

Indexing in XML

Due to a strong need for flexible interchange formats in a wide variety of web applications, XML is becoming extremely popular as a representation format for a wide variety of data, including business messaging, web site information integration, and catalog merging. However, since the type of XML data is not fixed, traditional queries are not always sufficient to locate the desired data. Consequently, regular expression queries are commonly supported by XML query languages. However, it is not clear that simple path access structures developed for Object-oriented databases will be suitable in this domain. In this project, we investigate practical, updateable index structures for regular expression queries on XML graph structures.
With Shriraghav Kaushik (Raghav), Ehud Gudes and Pradeep Shenoy .

Papers:

Main-Memory Databases

In particular, the DataBlitz Main-Memory Database (research name Dali). Please follow these links for more information on this project.

Papers: