From the course: Oracle Cloud Infrastructure AI Foundations

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

AI infrastructure overview

AI infrastructure overview - Oracle Cloud Infrastructure Tutorial

From the course: Oracle Cloud Infrastructure AI Foundations

Start my 1-month free trial Buy for my team

AI infrastructure overview

“

(gentle music) - [Instructor] Graphics Processing Unit, or GPU, is very important component of AI infrastructure. Let us see what is a GPU and why GPU is required. Machine learning and AI workloads need lots of repetitive calculations to be made on huge amount of data, especially for model training and inference as well. GPU is a piece of hardware that is incredibly good in certain type of calculations. Parallel computing on GPUs is designed to run many processes at the same time. GPU has thousands of lightweight cores all working on their share of data in parallel. This gives them the ability to crunch through extremely large datasets at tremendous speed. GPUs can process many inference tasks in parallel, leading to significantly higher throughput compared to CPUs. This is especially useful for batch processing or serving multiple requests simultaneously. Modern GPUs are optimized for deep learning frameworks like TensorFlow, PyTorch, and ONNX Runtime. These frameworks leverage…

Contents