PARA'04 State-of-the-Art
in Scientific Computing
June 20-23, 2004 (Home page)
Updated: 17 June 2004

High performance linear algebra algorithms; Session 2

Minisymposium organizer: Fred G. Gustavson

1. 16:00 - 16:20 Carsten Scholtes; Fachgruppe Informatik, Universitaet Bayreuth, Bayreuth, Germany
   "A Method to Derive the Cache Performance of Irregular Applications on Machines with Direct Mapped Caches" (slides)

2. 16:20 - 16:40 J. A. Gunnels, F. G. Gustavson, G. M. Henry, and R. A. Van De Geijn; IBM, IBM, Intel, and U. Texas(Austin), USA
   "Matrix Multiplication Kernels: Synergy Between Theory and Practice Leads to Superior Performance"

3. 16:40 - 17:00 J. A. Gunnels, F. G. Gustavson, K. Pingali, K. Yotov; IBM, IBM, U. Cornell, U. Cornell, USA
   "A General Purpose Compiler that obtains High Performance for Some Common Dense Linear Algebra Codes"

4. 17:00 - 17:20 David Barnes and Tim Hopkins; Computing Laboratory, University of Kent, Canterbury, Kent, UK
   "Applying Software Testing Metrics to Lapack"

5. 17:20 - 17:40 Peter Drakenberg; Royal Insttute of Technology, Kista, Sweden
   "A Matrix-type for Performance Portability"

Home page