Parallelization of Inference on GPUs for TFServing using Nginx

A guiding to load balancing your TFServing Inference API over multiple GPUs.