Title/Author/Abstract

• Title:
  Compiler and Middleware Support for Scalable Data Mining

• Author:
  Gagan Agrawal,Ruoming Jin, Renato Ferreira, and Xiaogang Li

• Full Paper(.ps version)

• Abstract:
  High performance data mining is emerging as an important class of parallel applications. The expertise and effort currently required in implementing, maintaining, and performance tuning a parallel data mining application is currently an impediment in the wide use of parallel computers for data mining.

  We have developed a data parallel dialect of Java that can be used for expressing common data mining algorithms at a high level. Our compiler generates a middleware specification from this dialect of Java. The middleware supports both distributed memory and shared memory parallelization, and performs a number of I/O optimizations to support efficient processing of disk resident datasets.

  In this paper, we describe the commonality between different data mining algorithms, the middleware and its interface, the data parallel dialect of Java, and the compilation techniques required for generating the middleware specification.

Home : Cpc2001