NVIDIA GPUs Scalability to Solve Multiple (Batch) Tridiagonal Systems Implementation of cuThomasBatch
URL: https://link.springer.com/chapter/10.1007/978-3-319-78024-5_22
Authors: Valero-Lara, Pedro / Martinez-Perez, Ivan / Sirvent, Raul / Martorell, Xavier / Pena, Antonio
Research Lines: Application optimization for GPU acceleration
Publication: Lecture Notes in Computer Science
Place Published: Springer
Volume / Pagination: 10777 / 243-253
Paraules clau: Tridiagonal linear systems, Scalability, Thomas algorithm, PCR, CR, parallel processing, cuSPARSE, CUDA