The publications below study two types of algorithms: stream algorithms and cache oblivious algorithms. Stream algorithms are a refinement of systolic algorithms that unleash the performance potential of superchips (stream architectures), an evolutionary step beyond the Raw architecture. Unfortunately, to date, the programming of stream algorithms is very cumbersome. Cache oblivious algorithms, on the contrary, are designed to provide portable performance with reasonable programming effort by minimizing the number of data transfers within a memory hierarchy without knowing implementation parameters like the cache size. Cache oblivious algorithms work well for applications that exhibit temporal locality.
- Matteo Frigo and Volker Strumpen, ``The Cache Complexity of Multithreaded Cache Oblivious Algorithms,'' 18th ACM Symposium on Parallelism in Algorithms and Architectures, August 2006, 271-280. Copyright © ACM
- Volker Strumpen and Matteo Frigo, ``Software Engineering Aspects of Cache Oblivious Stencil Computations,'' IBM Research Report RC24035, August 2006.
- Matteo Frigo and Volker Strumpen, ``The Memory Behavior of Cache Oblivious Stencil Computations (preprint),'' Journal of Supercomputing, 39(2), February 2007, 93-112.
- Matteo Frigo and Volker Strumpen, ``Cache Oblivious Stencil Computations,'' 19th ACM International Conference on Supercomputing, June 2005, 361-366. Copyright © ACM
- Volker Strumpen, Henry Hoffmann, and Anant Agarwal, ``Stream Algorithms and Architecture,'' Journal of Instruction-Level Parallelism 6, September 2004. (An earlier version is available as Technical Memo mit-lcs-tm-636, pdf.)
- Volker Strumpen, Henry Hoffmann, and Anant Agarwal, ``A Stream Algorithm for the SVD,'' (pdf) Technical Memo 641, Computer Science and Artificial Intelligence Laboratory, MIT, October 2003.
This section covers a mixed bag of topics related to computer architecture, networks, and operating systems.
- Volker Strumpen, ``Energy Efficiency of Parallel Computations under Voltage Scaling," 21st , October 10, 2013, 41-46. ( Workshop on Microelectronicspresentation slides)
- Volker Strumpen and Matteo Frigo, ``The Spiral Cache: A Self-Organizing Memory Architecture,'' IBM Research Report RC24767, March 2009.
- Volker Strumpen, ``A Multithreaded Processor Architecture with Implicit Granularity Adaptation,'' IBM Research Report RC24184, February 2007.
- Volker Strumpen and Arvind Krishnamurthy, ``A Collision Model for Randomized Routing in Fat-Tree Networks,'' Journal of Parallel and Distributed Computing 65(9), August 2005, 1007-1021. (An earlier version is available as Technical Memo mit-lcs-tm-629.)
- Volker Strumpen, ``R2P2--Raw Resource Allocation for PCA Processing, or, A Simple Operating System for Raw,'' Raw Memo 33, Computer Science and Artificial Intelligence Laboratory, MIT, June 2004.
- Michael Bedford Taylor, Jason Kim, Jason Miller, David Wentzlaff, Fae Ghodrat, Ben Greenwald, Henry Hoffmann, Jae-Wook Lee, Paul Johnson, Walter Lee, Albert Ma, Arvind Saraf, Mark Seneski, Nathan Shnidman, Volker Strumpen, Matt Frank, Saman Amarasinghe and Anant Agarwal, ``The Raw Microprocessor: A Computational Fabric for Software Circuits and General Purpose Programs,'' IEEE Micro 22(2), 25-35, March/April 2002. Copyright © IEEE
Distributed Systems/Grid Computing
This section offers studies on several key technologies for reliable parallel computing in clusters and the Internet, often called grid computing these days. An earlier set of results on parallel molecular sequence analysis was awarded the first prize of the Mannheim SuparCup ( Computerwoche Artikel in Deutsch) in 1993.
- Volker Strumpen, ``Portable and Fault-Tolerant Software Systems,'' IEEE Micro 18(5), 22-32, September/October 1998. Copyright © IEEE
- Balkrishna Ramkumar and Volker Strumpen, ``Portable Checkpointing for Heterogeneous Architectures,'' In 27th International Symposium on Fault-Tolerant Computing - Digest of Papers, Seattle, WA, 58-67, June 1997. (More details may be found in UI-ECE TR-96.6.1.)
- Volker Strumpen, ``Coupling Hundreds of Workstations for Parallel Molecular Sequence Analysis,'' Software - Practice and Experience 25(3) 1995, 291-304. (An earlier version is available as Technical Report ifi-93.28.)
- Volker Strumpen and Thomas L. Casavant, `` Exploiting Communication Latency Hiding for Parallel Network Computing: Model and Analysis,'' International Conference on Parallel and Distributed Systems, IEEE, Hsinchu, Taiwan, 622-627, December 1994. (More details may be found in ETH TR-216.)
- Clemens H. Cap and Volker Strumpen, ``Efficient Parallel Computing in Distributed Workstation Environments,'' Parallel Computing 19(11) 1993, 1221-1234. (An earlier version is available as Technical Report ifi-92.07.)