Management and Processing of Data Streams

Stanford University, Stanford CA

Investigators

Abstract

For many recent applications, the concept of a data stream, possibly infinite, is more appropriate than a data set. By nature, a stored data set is appropriate when significant portions of the data are queried again and again, and updates are small and/or relatively infrequent. In contrast, a data stream is appropriate when the data is changing constantly (often exclusively through insertions of new elements), and it is either unnecessary or impractical to operate on large portions of the data multiple times. The goal of this research project is to develop models and techniques for the management and processing of data streams. Sampling, summarization, and online approximation algorithms will be employed to facilitate query processing and data mining over streams. The results of this project will provide efficient data stream techniques for data management, memory management, query processing, data mining, and data analysis. In addition, a software prototype will be developed for experimentation with algorithms and query processing, and as a testbed for some sample applications of significant scope, such as networking monitoring and traffic engineering, and medical monitoring data.