Did you know Kubernetes can work with cameras, sensors, and USB devices?
Thanks to Akri, it can.
Akri is a CNCF project that makes edge devices, which are often too small to run Kubernetes, accessible as resources in your cluster. It detects devices like IP cameras or GPUs, connects them, and schedules workloads to use them.
Akri handles devices as they appear and disappear, so your workloads keep running smoothly.
If you want to know more or get involved go to https://docs.akri.sh/#CNCFCloud Native Computing Foundation (CNCF)#Kubernetes#EdgeComputing
Basically, at the edge, there's a heterogeneous ecosystem of devices, right? You have your server class devices, PC size devices, and your little IoT sensors. And with these IoT sensors, they often speak different protocols and they have different topologies, different security requirements. So Aubrey is like an abstraction layer that helps to address all of that. It reduces the manual configuration of onboarding those to your clusters at the edge so you can request your IoT. Leaf devices as resources, just as you would like CPU or memory, so you can use them in your custom applications. Join our slack #akari and Kubernetes and feel free to check out Project Audrey on GitHub.
Interesting! Is this an alternative to #k3s, which is also a CNCF Sandbox project?