Section 5.1 Program at a glance
All times are Central Daylight Time (CDT), in other words, the time in Texas.
Talks are scheduled in sessions. While there is 30 minutes scheduled for contributed talks, speakers should plan for a 20 minute talk and once discussion comes to an end, we move on to the next talk. Thus, the schedule is elastic.
Eventually Robert hopes to add recordings of the talks to these pages, with permission of the speaker.
Thursday September 26 | ||
8:30- | Zoom session, breakout rooms, and Discord are open for mingling. Coffee and muffins for in-person participants. |
|
(Click on "Breakout room" at bottom of zoom window.) | ||
8:55-9:00 | Welcome | |
Robert van de Geijn | ||
9:00am - 11:00 | Session 1 | Moderator: Devangi Parikh |
Devin Matthews SMU |
Tentative: The state of BLIS 1.0 and 2.0 More Info 5.2.16 |
|
Bhaskar Nallani AMD |
LPGEMM Enhancements in AOCL BLAS More Info 5.2.11 |
|
Sridhar Govindaswamy AMD |
Close coupling of AOCL BLAS in AOCL LAPACK More Info 5.2.14 |
|
Nikoli Dryden LLNL |
A Distributed Multilinear Algebra Library for Deep Learning More Info 5.2.1 |
|
Additional discussion | ||
11:00am - 11:15am | Break | |
11:15am -12:45pm | Session 2 | Moderator: Ishna Satyarth |
Cem Bassoy Technical University of Hamburg |
Fast and layout-oblivious tensor-matrix multiplication with BLAS More Info 5.2.6 |
|
Elliott Binder Carnegie Mellon University |
FAST Attention for Small Tensors More Info 5.2.4 |
|
Upasana Sridhar Carnegie Mellon University |
Layer fusion with composable abstraction More Info 5.2.5 |
|
Additional discussion | ||
12:45pm - 1:45pm | Lunch | |
1:30pm | Group picture | |
1:45pm - 3:15pm | Session 3 | Moderator: Chao Yin |
Grace Dinh Cornell University |
Cost Estimation and Bounds for Sparse Kernels More Info 5.2.8 |
|
Evarist Fomenko Nvidia |
NVPL BLAS Architecture and Implementation Overview More Info 5.2.9 |
|
Thijs Steel KU Leuven |
Communication efficient application of sequences of rotations to a matrix More Info 5.2.10 |
|
Additional discussion | ||
3:15pm- 3:30pm | Break | |
3:30pm - 5:00pm | Session 4 | Moderator: Devin Matthews |
Ishna Satyarth Southern Methodist University |
LTLT Decomposition of a Skew-Symmetric Matrix - Derivation More Info 5.2.20 |
|
Chao Yin Southern Methodist University |
LTLT Decomposition of a Skew-Symmetric Matrix - High Performance Implementation More Info 5.2.19 |
|
Devangi Parikh/Greg Henry UT Austin and Intel |
Accuracy study of Cascading GEMM More Info 5.2.18 |
|
Additional discussion | ||
5:00pm | End of Day | |
Friday Sept. 27 | ||
8:00- | Zoom session, breakout rooms, and Discord are open for mingling. Coffee and muffins for in-person participants. |
|
(Click on "Breakout room" at bottom of zoom window.) | ||
8:30am - 10:20 | Session 5 | Moderator: Nallani Bhaskar |
Arnav Sharma AMD |
BLAS Extension APIs More Info 5.2.12 |
|
Eleni Vlachopoulou AMD |
CMake Build System in AOCL BLAS More Info 5.2.13 |
|
Stepan Nassyr Juelich Supercomputing Center |
Simulating Parameterized Kernels on Parameterized Architectures More Info 5.2.15 |
|
Carl Kwan UT Austin |
The Cholesky Factorization Theorem in ACL2 More Info 5.2.2 |
|
Additional discussion | ||
10:20am - 10:35am | Break | |
10:35am -12:05pm | Session 6 | Moderator: Angelika Schwarz |
Joe Dobson Arm |
Strategy Selection in the Arm Performance Libraries More Info 5.2.3 |
|
Jim Demmel University of California, Berkeley |
How to grade the accuracy of an implementation of the BLAS; Short update on Exception Handling More Info 5.2.7 |
|
Devin Matthews and Robert van de Geijn Southern Methodist University |
Vertical integration of the linear and multilinear software stack More Info 5.2.17 |
|
Additional discussion | ||
12:15pm-1:30pm | Another event needs room |