Platform engineering is the discipline of designing and building toolchains and workflows that enable self-service capabilities for software engineering organizations in the cloud-native era.

Platform Engineering

<@U04GX8WHMM5> reviewing the docs re: you support multi tenancy, how do you do that with isolation of workloads and applications?  Do you support Firecracker VM's on a physical host?

It is configurable, but by default we rely on (1) the isolation provided by the container runtime (Docker or Podman), (2) a combination of VXLAN and Wireguard for tenancy and application separated networking, and (3) per-tenant and host keying material for sercrets management. Through some configuration we do support alternate containerd backends like *<https://github.com/firecracker-microvm/firecracker-containerd|firecracker-containerd>*.

I see this works for a totally “on site” edge (did I get the nomenclature right?  Meaning the edge are pops deployed by the company)

I read briefly about ingress but don’t recall the details. What do you think of ngrok ingress as a means for ingress for edge compute?

How would you recommend connecting from a local client to the edge?  Over public internet?  Or since the client is controlled (it’s a game client) it should connect directly to the nearest edge through a tunnel?

Catching up here, on <https://platformengin-b0m7058.slack.com/archives/C04RLL5BQ4W/p1683519510114989?thread_ts=1683069604.676939&amp;cid=C04RLL5BQ4W> -&gt; Yes, we focus on “on-site” environments, i.e. where there if limited amounts of compute in many locations, so between the device edge (IoT, phones) and the regional hyperscaler/CDN edge. Naming is hard in all contexts, but the nomenclature around the edge is particularily fragmented IMHO. More on that topic here: <https://avassa.io/articles/defining-the-edge-of-edge-computing/>