Video CDN

Video CDN takes a source video, transcodes it into an adaptive bitrate ladder, and serves it as HLS through the CDN. It maintains its own library of videos and collections, managed under Manage Videos in the dashboard.

Getting video in

There are two ingest paths, and they produce identical results:

• Direct upload. Upload a source file from the dashboard. It is queued for encoding as soon as the upload completes.
• URL import. Give it a publicly reachable URL and it fetches the file itself. This is the usual way to bring in video that already lives somewhere else — including a file in one of your storage zones.

Either way, encoding is asynchronous. A video moves through queued and processing states before it is playable, and the dashboard reports progress per video.

What you get back

Encoding produces a set of renditions at different bitrates and resolutions, an HLS manifest that ties them together, and generated thumbnails. Players are handed the manifest URL and pick a rendition based on the viewer's bandwidth.

How this relates to Edge Storage

This is the question that trips people up most often, so plainly: Video CDN does not use an Edge Storage bucket as its origin, and does not automatically ingest what you put there. The two are separate products with separate storage. Uploading an MP4 to a storage zone stores an MP4 and nothing else happens.

To move a file from storage into your video library, use URL import and point it at the file's CDN URL. After encoding, the video is served from the video library — the original in your storage zone is just the source, and you can delete it if you no longer need it.

When to use storage instead

If you have already encoded your own HLS renditions and just want them served, you do not need Video CDN. Put the segments and manifests in a storage zone and serve them over its CDN hostname — see Edge Storage. Video CDN earns its place when you want the transcoding done for you.