CIS SIG-Theory Matrix multiplication: Genericity and Efficiency
Speaker: Brice Boyer, North Carolina State U.
Matrix multiplication: Genericity and Efficiency ?
Matrix multiplication with another matrix or a vector is the core routine of most computational linear algebra problems: providing such fast routines is crucial to us. We first detail some of our work for specific matrix multiplication routines. This work in general is done with the exact linear algebra library LinBox in mind. The design and the future of this library are modeled by concepts such as genericity, adaptivity and efficiency. We show techniques developed for combining them and provide timings. We will conclude with perspectives concerning the introduction of parallelism.
Friday, November 16, 2012 at 11:15am to 12:05pm
Smith Hall, Room 102A
Smith Hall, University of Delaware, Newark, DE 19716, USA
- Event Type
-
Lectures & Programs, Students, College of Engineering, Academics
- Calendar
- Group
- ENGR - Computer & Information Sciences
- Hashtag
-
#cis_sig-theory_matrix_multiplication_genericity_and_efficiency
- Contact Email
- Contact Name
-
David Saunders
- Contact Phone
-
302-831-6238
- Subscribe
Recent Activity
No recent activity