6th Annual Public Health Information Network Conference: Utah phGRID demonstration project: grid-enabling the Environmental Public Health Tracking Network

Utah phGRID demonstration project: grid-enabling the Environmental Public Health Tracking Network

Tuesday, August 26, 2008: 3:50 PM
International D
Catherine Staes, BSN, MPH, PhD , Dept of Biomedical Informatics, University of Utah, Salt Lake City, UT
Sam LeFevre, MS , Bureau of Epidemiology, Utah Department of Health, Salt Lake City, UT
Ron Price , Center for High Performance Computing, University of Utah, Salt Lake City, UT
Wu Xu, PhD , Bureau of Epidemiology, Utah Department of Health, Salt Lake City, UT
Jeffrey Duncan , Utah Office of Vital Records and Statistics, Utah Department of Health, Salt Lake City, UT
Walter Scott , Center for High Performance Computing, University of Utah, Salt Lake City, UT
John Contreras , Bureau of Epidemiology, Utah Department of Health, Salt Lake City, UT
Zhiwei Lui , Bureau of Epidemiology, Utah Department of Health, Salt Lake City, UT
Adi Gundlapalli, MD, PhD, MS , Internal Medicine, University of Utah, Salt Lake City, UT
Julio Facelli, PhD , University of Utah, Dept of Biomedical Informatics, Salt Lake City, UT
Grid architecture is a promising methodology to aggregate and analyze disparate, heterogeneous data existing under independent administrative domains and provide computational and analytic resources on demand. There is a need to evaluate the feasibility and utility of the phGrid to address public health goals. We will describe a Utah Center of Excellence in Public Health Informatics project to prototype a Utah Public Health Grid to integrate data, analytic and reporting services. This prototype uses de-identified birth records from the Utah Department of Health vital records system and air quality data from the Department of Environmental Quality. The databases have been implemented in Oracle and MySQL, respectively and run in two different machines under separate administrative authority. We have embedded the databases into two independent WSRF containers. We implemented basic analytic services in a third system. The analytical services issue the necessary queries to the databases, perform the desired analysis locally and on the NSF TeraGrid or HPC clusters at Utah, and display the output in a desired workstation. This demonstration project addresses processes that are currently implemented in the Utah Environmental Public Health Tracking Network, however the current infrastructure for analyzing this data is not scalable and dynamic. We expect that the data required for these analyses will remain in disparate, heterogenous databases in different enterprises. Grid technologies should address scalability issues. Once the system is developed and a security infrastructure is established, scaling up to include more data, over different time frames, will not require an exponential growth in effort.