Comparison of NVIDIA Metropolis Microservices Deployment: Jetson Orin vs. Enterprise GPUs

Hi everyone,

I’m currently exploring options for deploying NVIDIA Metropolis microservices and would like to understand the differences between deploying on Jetson Orin edge devices versus enterprise-grade GPUs.

Specifically, I’m interested in:

  1. The performance and capabilities of deploying Metropolis microservices on Jetson Orin edge devices.
  2. The advantages and use cases for deploying Metropolis microservices on enterprise-grade GPUs.
  3. Any considerations for choosing between Jetson Orin and enterprise GPUs for deploying Metropolis microservices.


Here is the guide for Metropolis Microservices: Metropolis Microservices | NVIDIA Developer
Some of workflow works on Jetson, some of the workflow works on dGPU.
Can you share your use case? So we can understand and suggest.

Clarification on dGPU: Is it a Physical Device or Available on Cloud?

You can deploy Metropolis Microservices on both a Physical Nvidia GPU or Nvidia GPU on Cloude.

So could you please clarify if enterprise GPU has physical GPU or cloud GPU?

Can you share your use case? So we can advice based on your use case.