Winter Research Awards

Publish Date: 02/16/2010

Congratulations to Drs. Fox, Butt, and Tilevich who have recently been awarded research funding.

 

Dr. Ed Fox along with VT colleagues PI- Randy Murch, and co-PIs Lynn Abbott, Michael Hsaio, and University of North Texas colleague Bruce Budowle were awarded funding through the National Institute for Justice for their proposal “Establishing the Quantitative Basis for Sufficiency:  Thresholds and Metrics for Friction Ridge Pattern Detail Quality and the Foundation for a Standard”

 

This research award is to 'develop a quantitative approach to measuring and establishing a standard for 'sufficiency' of information available in friction ridge (fingerprint) patterns.  Sufficiency refers to the quality and quantity of visual information available to make comparisons between friction ridge patterns of questioned and known origins, and then identifications of the individual who provided the samples. Machine-aided systems routinely digitize and automatically compare finger- and palm prints that contain a sufficient amount and quality of information based on algorithms for that purpose. Digitized prints of poor quality require the intervention of human experts to perform comparisons and identifications. Latent fingerprints, such as those left at a crime scene which are often partial or distorted, are analyzed and compared by human experts, whose expertise and judgment is based on their training and experience. By convention, the worldwide fingerprint community does not use any quantitative standard to determine the quantity and quality of information in an image or for the number of points of comparison required to make identification.

 

Dr. Eli Tilevich has been awarded funding through an IBM X10 Innovation Grant Award for his proposal entitled: “Automatic Adaptation of Java Frameworks for X10 to Improve Programmer Productivity”.  This grant program is directed to academic research and curricular development at scale on cloud computing platforms based on the X10 programming language.  The X10 Innovation Awards program represents an ideal opportunity for academic partners to learn, teach and help develop an eco-system built around a modern, open-source parallel language suitable for developing applications for parallel architectures.

 

X10 is a new programming language, being developed at IBM Research, whose design goal is to improve development productivity for parallel applications. To become suitable for constructing the cloud applications of the future, X10 must also provide facilities to systematically express essential non-functional concerns, including persistence, transactions, distribution, and security.  In a realistic parallel application, non-functional concerns may constitute intrinsic functionality, but implementing them in X10 can be non-trivial. To streamline the implementation of non-functional concerns, the Java development ecosystem features standardized, customizable, and reusable abstractions called frameworks. Java frameworks are a result of a concerted, cooperative, and multi-year effort of multiple stakeholders in the Java technology, which has been tested and proven effective by billions of lines of production code. This project will explore how Java frameworks can be automatically adapted for use by X10 programmers. The proposed approach will help avoid duplicating the effort expended on creating Java frameworks to provide equivalent facilities for X10. By enabling X10 programmers to leverage the collective expertise of the developers and users of Java frameworks, this project aims at improving programmer productivity.

 

 Dr. Ali Butt also received research funding from IBM through the IBM Shared University Research Award.  His proposal was entitled  “A Simulation-based Toolkit for Designing MapReduce Clusters for Enabling Customized Cloud Computing.” The proposal is in collaboration with researchers at the IBM Almaden Research center and will work on developing simulation-based tools to enable researchers and IT practitioners to profile their MapReduce setups.

 MapReduce has emerged as a model of choice for supporting modern data-intensive applications, especially as an enable for cloud computing in realizing a robust, flexible paradigm for supporting high-level management amd execution of large-scale enterprise applictions, as well as for providing on-demand automatic scaling to a large number of compute, storage, and networking resources. Setting up and operating a large MapReduce cluster entails careful evaluation of various design choices and run-time parameters to acheive high efficiency. However, this design space has not be- en explored in detail. The proposed research will explore this design space, specifically, how vari- ous design choices and run-time parameters of a MapReduce system affect application performan- ce of MapReduce setups, by capturing such aspects of these setups as node, rack, and network configurations, disk parameters, and performance, data layout and application I/O characteristics, among others, and using this information to predict expected application performance.