An AI Edge Computing Cluster with Kubernetes and NVIDIA Jetson Nano Boards
Many enterprises might prefer to run AI models locally – on edge devices – but end up running in the cloud. Lack of compute power at scale, access to GPU resources, access to know how, and lack of storage are some of the roadblocks.
Advances in edge devices – including GPU technologies – in conjunction with containerization and orchestration using Kubernetes, have been chipping away at these challenges. We’re either close or already at a tipping point – maybe delayed only by – you guessed it – supply chain issues!!!!
