Primal Dual Affine Scaling on GPUs
This work addresses the need for faster linear optimization on GPUs, but the lack of quantitative results makes it an incremental contribution.
The paper presents a GPU-based implementation of the Primal-Dual Affine scaling method for linear optimization, introducing a CUDA-friendly technique to solve the symmetric positive definite subsystem and strategies to reduce memory transfer and storage. No concrete performance numbers are provided.
Here we present an implementation of Primal-Dual Affine scaling method to solve linear optimization problem on GPU based systems. Strategies to convert the system generated by complementary slackness theorem into a symmetric system are given. A new CUDA friendly technique to solve the resulting symmetric positive definite subsystem is also developed. Various strategies to reduce the memory transfer and storage requirements were also explored.