61. PALMS-CI: AN EMERGING CYBERINFRASTRUCTURE RESEARCH PLATFORM

Department: Computer Science & Engineering
Research Institute Affiliation: California Institute for Telecommunications and Information Technology (Calit2)
Faculty Advisor(s): Ingolf Krueger | William Griswold

Primary Student
Name: Barry Demchak
Email: bdemchak@ucsd.edu
Phone: 858-452-8700
Grad Year: 2010

Abstract
A primary mission of the exposure biology community is to understand where activity-related energy expenditure occurs as a function of time and space. So far, it has invested in the development of wearable devices that measure location and motion, leaving much of the data management and analysis for future work. At the same time, the NIH has required that researchers make their data available for use by other researchers, and HIPAA regulations place significant restrictions on how data can be shared. Finally, as wearable and cell phone-based sensors proliferate, significant opportunities arise for large scale and continuous collection, analysis, and leveraging of personal data.

The PALMS (Physical Activity Location Measurement System) project is developing a cyberinfrastructure (PALMS-CI) as a highly scalable, policy-driven system capable of managing organizational and observational data for a number of studies and ultimately facilitating NIH, HIPAA, and other policy objectives. Researchers will define the data maintained for each of their studies, the devices used to collect the data, and the particular protocols used to analyze and visualize the results. In providing a consistent and common infrastructure, the PALMS-CI will enable policy-based sharing and reuse of data, device interfaces, calculations, protocols, analyses, and visualizations, thereby enabling and incentivizing the participation of a large number of stakeholders, including researchers, study participants, funding organizations, regulatory agencies, and groups presently unknown.

The PALMS-CI leverages Model-Driven Architecture (MDA) techniques to create a Service-Oriented Architecture (SOA) realized as services on an Enterprise Service Bus (ESB). Using the Rich Services development methodology, stakeholders are identified, requirements and use cases are collected, and a domain model is created – all of which support service definitions organized in a hierarchy according to a Rich Services pattern. An agile development process is used to continuously discover and evaluate requirements and prioritize changes, thereby defining the emergent feature set of the PALMS-CI and managing its development.

To date, the PALMS-CI is capable of defining and maintaining study and subject data, storing and analyzing observation data, and depicting correlations on geo-displays. Its service orientation and ESB-based implementation provide for exceptional flexibility in accommodating new and changing requirements, and they create a research platform for emerging technologies such as policy design, decision, and evaluation frameworks; dynamic service binding; service-oriented modeling; transaction modeling in loosely coupled systems; storage virtualization; service-oriented validation and verification; large scale data acquisition and continuous analysis; health monitoring and modification; cloud computing; and research-oriented social networks.

« Back to Posters or Search Results