BLIS Retreat 2017
Program
Contributed talksMonday Sept. 18
| Morning | GDC 6.302 | ||
| 8:30-9:00 | Breakfast (GDC 6.302) | various breakfast tacos | |
| 9:00-9:10 | Welcome | Robert van de Geijn | |
| 9:10-9:30 | Induced methods for complex matrix multiplication | SLIDES | Field Van Zee |
| 9:30-9:50 | High Performance Computing Kernels in N-body Problems | SLIDES | Chenhan Yu |
| 9:50-10:10 | Coffee Break | ||
| 10:10-10:30 | Strassen's Algorithm for Tensor Contraction | SLIDES | Jianyu Huang |
| 10:30-10:50 | Performance portable line smoother for multiphysics problems using compact batched BLAS |
NO VIDEO SLIDES |
Kyungjoo Kim |
| 10:50-11:10 | Coffee Break | ||
| 11:10-11:30 | Open BLAS G2 Standardization Issues | Greg Henry | |
| 11:30-11:50 | Mixing Flavors using BLIS | SLIDES | Devangi Parikh |
11:50 - 1 Lunch (GDC 6.302 - Computer Science Faculty Lounge) Barbeque and vegetarian options.
| Afternoon | GDC 6.302 | ||
| 1:00-1:20 | I Don't Care About BLAS |
VIDEO
SLIDES |
Devin Matthews |
| 1:20-1:40 | Leveraging modern supercomputing infrastructure for tensor contractions in large electronic structure calculations |
VIDEO (contact speaker) SLIDES |
Ilya Kaliman |
| 1:40-2:00 | Coffee Break | ||
| 2:00-2:20 | Shared-Memory Parallelization of MTTKRP for Dense Tensors | SLIDES | Koby Hayashi |
| 2:20-2:40 | Parallel 3D Cholesky-QR2 for square and rectangular matrices | VIDEO | Edward Hutter |
| 2:40-3:00 | Coffee Break | ||
| 3:00-3:20 | Overreaching with Outreach: Is HPC in Reach? | SLIDES | Maggie Myers and Robert van de Geijn |
| 3:20-3:40 | A portable DMA support for Level-3 BLIS on DMA-enabled embedded hardware | SLIDES | Minh Quan |
| 3:40-4:00 | Coffee Break | ||
| 4:00-4:20 | BLIS optimizations for AMD EPYC(TM) Processors | SLIDES | Santanu Thangaraj |
| 4:20-4:40 | BLAS on AMD GPUs | Andrew Chapman | |
Tuesday Sept. 19
| Morning | GDC 6.302 | ||
| 8:30-9:00 | Breakfast (GDC 6.302) | ||
| 9:00-9:20 | The MOMMS Family of Algorithms for Classical Matrix-Matrix Multiplication for Hierarchical Memory Architectures | SLIDES | Tyler Smith |
| 9:20-9:40 | BLASFEO: basic linear algebra subroutines for embedded optimization |
VIDEO SLIDES PAPER |
Gianluca Frison |
| 9:40-10:00 | Coffee Break | ||
| 10:00-10:20 | Integer GEMM (under) performance | SLIDES | Marat Dukhan |
| 10:20-10:40 | Leveraging Dense Linear Algebra Performance for Graphs, Stencils and Sparse Linear Algebra | Richard Veras | |
| 10:40-11:00 | Coffee Break | ||
| 11:00-11:20 | Packing - the next BLIS frontier? | SLIDES | Tze Meng Low |
| 11:20-11:40 | BLISful Wrapup | Robert van de Geijn | |