GPU-enabled steady-state solution of large Markov models

We describe a novel parallel steady-state solver that uses NVIDIA's Compute Unified Device Architecture (CUDA) library to perform calculations on a graphics processing unit (GPU). We demonstrate speed-ups of over 8 times compared with a CPU-only solver. We also discuss a parallel implementation which runs on multiple GPUs on separate machines, and explain how we deal with allocating appropriate amounts of work to heterogeneous computing resources