[LA-SiGMA-TESC] Technologies for Extreme Scale Computing seminar on Thursday (3/26)

[Wei P Feinstein]

Hi all,

We will continue our weekly TESC seminar this Thursday (3/26) at 12:00pm-1:00pm in 1034 Digital Media Center.

Rakib Hasan, a graduate student from the division of computer science and engineering, will present his research. The title is "Build your personalized Matrix-multiply kernels using ATLAS:

Abstract:
GEneral Matrix-matrix Multiply (GEMM) is one the most widely used linear algebra routines.  But the conventional use of GEMM through the BLAS interface contains overheads that can only be amortized for large operand (matrix) sizes.  Moreover, it may unnecessarily pollute the cache. The new microkernel tuning framework of ATLAS allows you, without understanding too much details of GEMM optimizations, to implement your own personalized GEMM that can help avoid such overheads for small-to-moderate operand sizes and to explicitly manage the cache. The drawback, of course, is that you need to understand some details of GEMM implementation instead of using the convenient BLAS interface.  In this talk, I will describe how you can autotune the microkernels to build your own personalized GEMM.

PIZZA AND DRINK WILL BE PROVIDED.
Presentation time: 12:00PM~1:00PM CST
Location: 1034 Digital Media Center


The meeting will be shared through adobe connect at http://connect.lsu.edu/gpu/


Best,
Wei

Wei Feinstein, Ph.D.
High Performance Computing
Louisiana State University
E-mail: wfeinstein at lsu.edu
Office: 337 Frey Computing Services Center
Tel: 225-578-0582
--------------------------------------------------
LASiGMA-gpu mailing list
LASiGMA-gpu at loni.org<mailto:LASiGMA-gpu at loni.org>
https://mail.loni.org/mailman/listinfo/lasigma-gpu
wiki: https://wiki.cct.lsu.edu/la-sigma/
http://www.institute.loni.org/lasigma/lasigma-gpu.php


-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mail.loni.org/mailman/private/lasigma-gpu/attachments/20150324/cd8ede67/attachment.html