Hosting » AWS » What is GPU instance in AWS?

What is GPU instance in AWS?

Last updated on September 25, 2022 @ 5:56 pm

GPU instance refers to AWS cloud computing instances that uses graphics processing unit (GPU) cores. They are designed for high-performance computing (HPC) and graphics applications.

AWS offers two types of GPU instances: the General Purpose GPU Instance and the Compute Engine Flexible GPU Instance.

PRO TIP: GPU instances in AWS are not to be used for general purpose computing. They are designed for specific workloads that can take advantage of the GPU’s parallel processing power. Attempting to use a GPU instance for general purpose computing will likely result in slow performance and increased costs.

AWS General Purpose GPU Instance is designed for general purpose computing and is optimized for throughput and low latency. It features 8GB of on-board memory and supports up to four NVIDIA K80 GPUs.

AWS Compute Engine Flexible GPU Instance is designed for high-performance computing (HPC) and graphics applications. It offers the same capabilities as the General Purpose GPU Instance, but it is also optimized for flexibility.

It features 3GB of on-board memory and supports up to two NVIDIA K80 GPUs.

Drew Clemente

Drew Clemente

Devops & Sysadmin engineer. I basically build infrastructure online.