numerical linear algebra

ISC21 Workshop: Numerical Algorithms and Libraries for Exascale (NAL-X)

-

ISC21 KAUST Frankfurt Germany Time CET

Abstract With the hardware technology scaling and the trend on heterogeneous chip design, the existing numerical algorithms and software framework may break down due to load imbalance. There is currently a fundamental mismatch between the underlying hardware architecture with high thread concurrency and the software deployment of numerical libraries, which relies on the traditional bulk synchronous programming model. Numerical software should first squeeze performance out of single node by efficiently running on manycore architectures with processor counts sharing a common memory in the

Heterogeneity in Hardware: Opportunities and Challenges for Software and Applications (SC21 Panel)

-

SC21 Saint Louis MO USA

Heterogeneity in Hardware: Opportunities and Challenges for Software and Applications SC21 Panel Time: Tuesday, 16 November 2021, 10:30am - 12pm CST Location: 225-226 (America's Center, St. Louis, MO USA) Abstract With the end of Moore’s Law, the community has witnessed new hardware trends to increase performance. Today, it is not only the traditional x86 and accelerators that are part of computing systems, but also ARM, FPGAs and dedicated processors for DL workloads that equip now pioneering HPC systems. By the end of this decade, we are moving towards an era of extreme scale with “extreme

KAUST scientists win prestigious Gauss Award for supercomputing excellence

1 min read ·
KAUST Ph.D. graduate Dr. Noha Al-Harthi and doctoral student Rabab Alomairy, have won the German Gauss Center for Supercomputing (GCS) Award for original research that best advances high-performance computing. This makes KAUST the first Middle Eastern institution to receive this prestigious award.

H2Opus: a Performance-Oriented Library for Hierarchical Matrices

1 min read ·
H2Opus: a Performance-Oriented Library for Hierarchical Matrices Download at https://github.com/ecrc/h2opus. Experience our latest HPC software release that implements H 2-Matrix operations on shared-memory systems, possibly equipped with GPU hardware accelerators. The prime target applications for H2Opus are PDE-constrained optimizations. The features of H2Opus include: Generation of matrix structure from a point set and admissibility condition, Construction of a hierarchical matrix given a kernel function, Matrix-vector and matrix-multiple-vector multiplication, Basis orthogonalization

Our Masked Guest Speakers Series

1 min read ·
Every week, we have a masked guest speaker opening our HiCMA weekly meeting by giving a remote talk on broad topics related to numerical linear algebra, high performance computing, programming models and performance optimizations. Our motto is fun and science!

Steering Customized AI Architectures for HPC Scientific Workloads: A Bird of Feather at SC22!

1 min read ·
Abstract: We explore the possibilities of a hybrid system capable of solving both HPC and AI scientific problems. Such a hybrid architecture combines the synergism between classical HPC platforms and dedicated AI chip systems, which is important due to the computational challenges brought to the fore by massively parallel Exascale systems. We discuss the system functionality, the algorithmic software adaptations, and performance considerations. We present efforts in supporting AI/ML applications in addition to seismic imaging, climate/weather prediction, and computational astronomy on hybrid