Compile-Time Sparse Matrices - Optimizing Memory and Performance
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only! Grab it Explore compile-time sparse matrices in C++ linear algebra libraries through this conference talk from Meeting C++ 2023. Delve into the limitations of classical libraries that provide dense matrices and sparse matrices with run-time sparseness. Learn how to incorporate sparseness information into matrix classes at compile-time, reducing memory overhead while maintaining run-time efficiency. Discover techniques for storing only necessary entries, making matrices memory-efficient. Understand how this approach enables automatic removal of unnecessary operations by the compiler, resulting in performance comparable to or even surpassing hand-written sparse-matrix code. Gain insights into combining memory efficiency with run-time efficiency for a "free lunch" in C++ linear algebra implementations.