Acceleration of Commercial Low-Rank Matrix Solver via SLATE

Performing accurate simulations of large- and multi-scale electromagnetics problems has far-reaching implications The same physics governs applications as diverse as rare earth content reduction in DC motor design; wireless propagation in dense urban environments and novel antenna design, both key issues in the pending 5G communications boom; electromagnetic sounding for subsurface exo-terrestrial interrogation; burgeoning RF-based medical imaging modalities and therapies; and exploration of intra- and inter-cellular quasistatic and RF signaling With such diversity of the simulation problems, a scalable, high performance computational electromagnetics (CEM) software package will provide state-of-the-art design and analysis capability to commercial, medical, and government users This program will extend an already-commercialized surface integral equation CEM software package to include distributed solution capability to achieve massive increases in simulation performance The current software supports multi-core and multi-GPU capability and will now incorporate a Department of Energy-funded exascale linear algebra software suite to address extremely large-scale electromagnetics design and analysis problems Importantly, this work will also extend the linear algebra suite to support direct solution of matrices compressed via low-rank/H-matrix exploiting algorithms In addition to developing new capabilities for a commercial CEM software package, the project will extend the linear algebra package’s support for low-rank algorithms, an important recent area of development in computational physics This work has two major objectives First, the team will demonstrate the feasibility and performance scaling of the current software extended to support the DOE linear algebra package addressing full-rank matrices This will provide confidence toward the ultimate goal of a low-rank-exploiting linear algebra engine Additionally, this objective will provide necessary data for validating large-scale numerical results currently not achievable with the single-node commercial version of the software Second, the team will demonstrate an initial capability for the low-rank compressed matrix solution using several real-world examples relevant to the small business’s other technical focuses By achieving this objective, the team will have the baseline capability available for extension and performance optimization in Phase II This new software will provide an important leap in computational capability for all users requiring electromagnetic design and analysis capability We anticipate these improvements to be particularly useful within the 5G communications and medical imaging communities These new capabilities will also enable the small business to provide high performance results using a simulation-as-a-service business model since not all users may have access to large-scale computational resources