Learn about the performance advantages of High Bandwidth Memory (HBM) and near-memory processing in sparse multi-physics applications through a 16-minute technical talk from Los Alamos National Laboratory scientist Jered Dominguez-Trujillo. Discover how the national lab addresses the growing gap between compute throughput and memory bandwidth by implementing HBM on CPU nodes to enhance workload performance. Gain insights into their forward-looking research partnerships with industry and academia, exploring novel near-memory architectures and the development of evaluation tools, benchmarks, and models for these emerging technologies. Understand the implications of Moore's Law ending and Dennard Scaling on computational performance, particularly for codes requiring indirect memory access patterns.