Compute Canada

View All Papers
To Submit a new publication, Please visit the User's Section (logged in users only)

Publications:

To view a publication's Abstract (if available), click on publication's title

Author(s)

     F.Dehne, G.Hickey, A.Rau-Chaplin, M.Byrne

Title

     Parallel catastrophe modelling on a Cell B.E.

Publication Name

     Int. Journal of Parallel, Emergent and Distributed Systems (formerly Parallel Algorithms and Applications)

Publication Type

     Journal
Publisher
     
Place
     
Editor(s)
     
Number
      5
School
     
Volume
      25
Chapter
     
Pages
      401-410
Other Publication Information
     
Publication Year
     2010
Abstract
     In this paper, we study the potential performance improvements for catastrophe (CAT) modelling systems that can be achieved through parallelisation on a cell processor [Cell Broadband Engine Architecture (Cell/BE)]. We studied and parallelised a critical section of CAT modelling, the so-called inner loop and implemented it on a Cell/BE running on a regular Playstation 3 platform. The Cell/BE is known to be a challenging environment for software development. In particular, the small internal storage available at each synergistic processing element (SPE) of the Cell/BE is a considerable challenge for CAT modelling because the CAT-modelling algorithm requires frequent accesses to large lookup tables. Our parallel solution is a combination of multiple techniques: streaming data to the SPEs and parallelising inner loop computations, building caches on the SPEs to store parts of the large CAT modelling lookup tables, vectorising the computation on the SPEs and double-buffering the file I/O. On a (Playstation 3) Cell/BE with six active SPEs and four-way vectorisation on each SPE, we were able to measure a sustained 16 speedup for our parallel CAT-modelling code over a wide range of data sizes for real-life Japanese earthquake data.
Keywords
     parallel algorithms, catastrophe modelling, cell processor
Website
     http://www.tandfonline.com/doi/abs/10.1080/17445760903492086
Paper Status
     Published
DOI / Publication ID
     10.1080/174457609034



View All Papers