Computer & Information Science Department   Polytechnic University

ATTENTION: THIS WEB SITE HAS MOVED. The pages you are looking at are no longer being maintained. Please go to http://www.poly.edu/cis/ to visit the new site of the Department of Computer and Information Science at Polytechnic University.

Exploiting Coarse-Grained Parallelism to Accelerate Protein Motif Finding with a Network Processor

Patrick Crowley
Washington University in St. Louis


Tuesday, Oct. 25, 11:00am
LC 400, Brooklyn Campus, Polytechnic University

 

Abstract      

        While general-purpose processors have only recently employed chip multiprocessor (CMP) architectures, network processors (NPs) have used heterogeneous multi-core architectures since the late 1990s. NPs differ qualitatively from workstation and server CMPs in that they replicate many simple, highly efficient processor cores on a chip, rather than a small number of sophisticated superscalar CPUs. In this talk, we compare the performance of one such NP, the Intel IXP 2850, to that of the Intel Pentium 4 when executing a scientific computing workload with a high degree of thread-level parallelism. Our target program, HMMer, is a bioinformatics tool that identifies conserved motifs in protein sequences. HMMer represents motifs as hidden Markov models (HMMs) and spends most of its time executing the well-known Viterbi algorithm to align proteins to these models. Our observations of HMMer on the IXP are therefore relevant to computations in many other domains that rely on the Viterbi algorithm. We show that the IXP achieves a speedup of 1.82 over the Pentium, despite the Pentium's 1.85x faster clock. Moreover, we argue that next-generation IXP NPs will likely provide a 10-20x speedup for our workload over the IXP 2850, in contrast to 5-10x speedup expected from a next-generation Pentium-based CMP.

Bio

      Patrick Crowley is assistant professor in the Department of Computer Science and Engineering at Washington University in St. Louis, where he is a member of the Applied Research Laboratory. Crowley's current work has three main areas: A) the design and application of multi-core processors for networking and other areas such as computational biology and scalable virtual machine implementations, B) the design of a storage-based supercomputer-class system that exploits advanced FPGAs and multi-core processors to scale scientific compute throughput with the rate of data transfer off a local disk, and C) high-performance router architectures, with a particular emphasis on the design of diversified routers capable of supporting multiple coexisting network architectures on a single provisioned physical substrate. Crowley is co-editor of the three-book series "Network Processor Design" and co-founder of the ACM/IEEE Symposium on Architectures for Networking and Communications Systems (ANCS), the first of which will be held in Princeton, N.J. in October 2005. More information is available on his homepage.

      For further information please contact Haldun Hadimioglu (haldun at photon.poly.edu)