Henry (Hank) Hoffmann

Email: hank at mit dot edu


I am a Ph.D. student at MIT's Computer Science and Artificial Intelligence Laboratory where I am a member of the Raw group.   I was formerly an associate technical staff member in MIT Lincoln Laboratory's Embedded Digital Systems Group.  My research interests include parallel algorithms and architecture and automated parallelization techniques.

My resume is available here.


Research

Stream Algorithms:

Along with Volker Strumpen and Anant Agarwal, I have developed a framework for analyzing the computation, area, and energy efficiency of computations on single-chip parallel architectures.  This work has led to extremely efficient, scalable implementations of several difficult to parallelize algorithms, including QR factorization and singular-value decomposition.

Embedded Parallel Computing:

As a member of MIT Lincoln Laboratory's Embedded Digital Systems Group, I co-designed a library for scalable, portable, parallel digital signal processing.  I also worked on algorithms and software systems that would automatically parallelize digital signal processing applications.