A Short Note on Gaussian Process Modeling for Large Datasets using Graphics Processing Units
This work addresses computational bottlenecks for researchers using Gaussian process models in statistical applications, but it is incremental as it applies existing GPU acceleration methods to this specific domain.
The study tackled the high computational cost of fitting Gaussian process models for large datasets by implementing them on a CPU+GPU heterogeneous system, resulting in a significant reduction in computational cost compared to traditional CPU-only systems.
The graphics processing unit (GPU) has emerged as a powerful and cost effective processor for general performance computing. GPUs are capable of an order of magnitude more floating-point operations per second as compared to modern central processing units (CPUs), and thus provide a great deal of promise for computationally intensive statistical applications. Fitting complex statistical models with a large number of parameters and/or for large datasets is often very computationally expensive. In this study, we focus on Gaussian process (GP) models -- statistical models commonly used for emulating expensive computer simulators. We demonstrate that the computational cost of implementing GP models can be significantly reduced by using a CPU+GPU heterogeneous computing system over an analogous implementation on a traditional computing system with no GPU acceleration. Our small study suggests that GP models are fertile ground for further implementation on CPU+GPU systems.