Edge where the subscribers are
Edge nodes placed against your topology and peering reality, so the common case is served from inside the network and transit carries only the misses.
Delivery that stays inside your own network
If you already own the network the viewer is on, paying a third party to deliver video across it is paying twice. In-network delivery is the clearest case there is for owning the edge — the traffic never has to leave.
Who this is for
Operators who already carry the traffic and would rather terminate it inside their own network than pay to fetch it from outside.
- Telcos
- Mobile and fixed-line operators launching or operating a video service for their own subscriber base.
- Internet service providers
- Networks where video is already the dominant traffic class and transit is the line item it drives.
- Enterprise network operators
- Organisations running their own AS and address space who intend to deliver from inside it.
The traffic is already on your network
For an operator, video is usually the largest single component of transit — and every gigabyte of it is content your own subscribers are pulling in from outside. Placing the edge inside the network turns an external transit cost into an internal one, and shortens the path at the same time. Cheaper and faster is a rare combination; this is one of the places it genuinely exists.
The second advantage is that you already know who the viewer is. Subscriber identity, entitlement and tiering can be integrated properly rather than bolted on, which makes bundled video an actual product rather than a reseller arrangement.
What an operator deployment has to get right
Subscriber entitlement, properly integrated
Your subscriber system stays authoritative. The platform consumes identity and tier rather than maintaining a second, competing customer record.
Capacity that follows the network
Adding an edge site is adding a declared node, not a project. The estate is Terraform, so a new point of presence is a merge request and a rack visit.
The platform is yours, not a vendor's
No third party sits between you and your subscribers' viewing data, and no contract renewal decides whether you can keep operating your own service.
The usual operator shape
Application tier wherever it is convenient; delivery firmly inside your network.
Untrusted — the public internet
- Web player
- Mobile apps
- TV apps
- DNS
No media byte is ever served from here
- Load balancer
- Platform APIs
- session-authz
- Managed database
- Object storage
- CDN
Every media byte leaves from here
- traefik-edge
- CrowdSec
- Live path
- VOD path
The bytes never enter a metered path
- live-transcode
- live-origin
- vod-transcode + GPU
- vod-origin-cache
- MinIO
- key-delivery
- OpenBao
- ClickHouse
The same pipeline provisions AWS and the metal
- Proxmox VE
- Terraform
- Ansible
- Cost controls
- Prometheus
- Grafana
The things operators ask first
Can we run this as a white-label service for our subscribers?
Yes — the viewer-facing apps and web front end are branded to you, and subscriber identity comes from your systems.
How many edge sites do we need?
That is a topology question rather than a platform one. It follows where your subscribers concentrate and where your peering already is, and it is one of the outputs of discovery.
Does this work alongside a commercial CDN?
Yes, and that is often the sensible first step: serve the in-network baseline yourself and let the existing CDN handle overflow and out-of-network traffic.
Where does your video traffic exit today?
Transit volume, peering arrangements and subscriber count are enough to model what in-network delivery would change.