Research Interests
High-Performance Bioinformatic Computing
High performance data-parallel associative computing is used to approach a common bioinformatics application: sequence alignment.
The current focus is to return the top k optimum alignments instead of a single, best local alignment based on the rigorous Smith-Waterman algorithm. The results are comparable to the Eggert-Waterman alignment algorithm but it takes advantage of several properties of the massively parallel ASsociative Computing model, or ASC.
Associative Computer Model (ASC)
ASC (ASsociative Computing) is a data-parallel, or SIMD model with certain extensions that are not normally identified with SIMDs. ASC is a generalized version of the associative computing style that has been in use since the 1970’s with the introduction of associative SIMD computers such as the STARAN. Associative computing includes the use of data parallel programming, which is sequential in nature. Unlike MIMD programming, ASC programmers are not responsible for task allocation, load balancing, synchronization points, etc.
 |
More information about ASC is listed on the ASC reference page. |
Conceptual view of the ASC model |
Associative Smith-Waterman
Currently, optimizations are being made to the ASC emulator as well as the associative version of the Smith-Waterman algorithm. The emulator optimizations are being made to enable larger sequences in the test runs.
