Easier Heterogeneous Clustering

I want to expand my Olares cluster by connecting a low-power, ARM-based single-board computer (like an Orange Pi 5 Ultra or Raspberry Pi) to my Olares One. My goal is to keep the energy-sipping ARM board always on for lightweight tasks like Home Assistant, while putting the power-hungry Olares One to sleep at night and waking it up on-demand only for heavy AI workloads.

But right now, setting up a mixed x86/ARM cluster is way too complicated. The system doesn’t automatically handle the image architecture differences between amd64 and arm64, and running the core daemon across different platforms still requires a lot of manual, advanced Kubernetes hacking to prevent workloads from breaking.

I hope the team can look into resolving these cross-architecture compatibility gaps in future updates. It would be amazing to have seamless x86 + ARM hybrid clustering alongside some smart power-scheduling features, making Olares the ultimate efficient ecosystem for home lab enthusiasts!