Scaling Decision Optimization to 100 Million Variables and Beyond with mPDLP in NVIDIA cuOpt

NVIDIA cuOpt GPU-accelerated decision optimization can deliver speedups of more than 10x over CPU solvers on a single GPU for large-scale linear programming problems. cuOpt is now solving the zib03 benchmark, which has over 104 million nonzeros, on multiple GPUs with mPDLP, achieving a solve rate nearly 10x faster than a year before.