About Us People Research Databases Software Education Education Resources
 
     
 
»  HomePage
»  About Us
»  People
»  Research
»  Databases
»  Software
»  Publications
»  Education
»  Graduate Training
»  Course Web Sites
»  CCBM Seminars
»  9/23/2004 -Michael Beer
»  9/21/2004 -Myong-Hee Sung
»  8/18/2004 - Jason Papin
»  4/20/2004 - Tanveer Syeda-Mahmood
»  4/13/2004 - Laurent Younes
»  3/30/2004 - Michael Miller
»  3/23/2004 - Donald Geman
»  3/9/2004 - Joel Bader
»  2/18/2004 - Albert Montillo
»  2/10/2004 - Aik Choon Tan
»  12/11/2003 - Jennifer Van Eyk
»  11/06/2003 - Eduardo Marbán
»  10/23/2003 - Brian O'Rourke
»  9/12/2003 - Shankar Subramaniam
»  Links
»  Site Map
»  Login
»  User Registration
 

CCBM Seminar Series

Tanveer Syeda-Mahmood

IBM Almaden

MineLink: A Novel Distributed Computing Architecture for Seamless Data and Application Integration

With the advent of distributed databases, large-scale repositories, computational analysis and scientific visualization tools, the IT requirements for today's scientists are considerably more complex.  Such distributed resources have  opened up new possibilities for collaborative research for scientists  through sharing of  experimental data and data analysis tools to look for previously unseen correlations or relationships. Translating the above possibilities into reality, however, requires considerable effort on the part of both computational scientists and biomedical researchers to break down the barriers that exist between researchers and the access and analysis of their scientific data.  Among the important technology gaps are those that lie in the areas of  seamless data and application  integration (information integration) and design of ease-of-use systems.

MineLink is a system developed at IBM Almaden Research that addresses these gaps through a novel programmable framework for the automatic integration of services, be they data miners, visualization components, or databases starting from task-specific workflow requests from end-users.  

In this talk, I will describe the MineLink architecture and the underlying algorithms for seamless data and application integration through automatic service composition. Specifically, I will describe the approach used by MineLink   to physically compose a  given set of services by semantically matching schemas derived from the API specification of source and destination services through data  mining of their relationships. The algorithms in MineLink and its architecture are now influencing key software products in enterprise data and application integration by IBM

For information on disability access contact Anne Albinak at 410-516-5310 or aalbinak@bme.jhu.edu




Tuesday, April 20, 2004
4:00-5:00pm

Room 110,
Clark Hall

and

videocast to

Talbot Library,

709 Traylor Building