1 / 3

Performance of Graph and Biological Analytics on the IBM Cell Broadband Engine Processor

Performance of Graph and Biological Analytics on the IBM Cell Broadband Engine Processor. David A. Bader Tan M. Tran Georgia Institute of Technology. Biological Analytic: SSCA #1. Background and Intent.

nerice
Télécharger la présentation

Performance of Graph and Biological Analytics on the IBM Cell Broadband Engine Processor

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Performance of Graph and Biological Analytics on the IBM Cell Broadband Engine Processor David A. Bader Tan M. Tran Georgia Institute of Technology

  2. Biological Analytic: SSCA #1 Background and Intent • To develop analytical schemes to identify similarities between sequences of symbols to assist computational biologists • Optimize for the IBM Cell Broadband Engine Processor • Find common genes and closely matches between sequences • Five kernels • Methods: • Local alignment: Water Smith dynamic programming (Kernel 1) • Searching for similarities (Kernels 2 and 3) • Global alignment (Kernel 4) • Multiple alignment: Center Star method (Kernel 5) • Each kernel operates on either the original sequences, the results of the previous kernel, or both • To be entirely integer and character based

  3. Graph Analytic: SSCA #2 Background and Intent • To develop a scalable synthetic compact application that consists of four kernels requiring irregular access to a large, directed, weight multi-graph • Optimize for the IBM Cell Broadband Engine Processor • Untimed Scalable Data Generator generates the graph as tuples of vertex pairs and corresponding weights. • Methods: • Cache-friendly adjacency lists (Kernel 1) • Parallel scan to classify larger sets of graph (Kernel 2) • Parallel BFS (Kernel 3) • Dijkstra Single-Source Shortest Path and Betweenness Centrality algorithms(Kernel 4) • David A. Bader and KameshMadduri designed and implemented the first parallel Betweenness Centrality algorithm on symmetric multiprocessors. • To be entirely integer and character based

More Related