From the moment it leaves your machine to the moment somebody in another timezone finishes downloading it.
Your browser talks to the closest ingest point rather than to a single origin. A 40 GB master leaving Lisbon does not travel through Frankfurt first.
Pools are flat — no folder tree to design. Each pool has its own members and its own retention window.
Expiry date, download cap, optional password. Defaults come from the pool, so nobody has to remember them for each file.
The first request in a region warms a cache; everyone after that is served locally. The download page is plain HTML — it opens on a locked-down office laptop.
One link per job or client. It points at a pool and carries the same expiry rules as an outbound link.
The page asks for a name and the files. Nothing to install, nothing to register, no password for them to lose.
Transfers pick up where the connection dropped. A large upload survives a train tunnel or a hotel network.
Who uploaded what, when, from where and how big it was — in the pool log, and by email if the pool is set that way.
The numbers people ask about on the first call, in one place.
| Largest single file | 2 TB |
| Files per transfer | no limit |
| Link lifetime | 1 hour to 365 days |
| Download cap per link | 1 to unlimited |
| Resume window after a dropped upload | 48 hours |
| Storage regions | Frankfurt, Amsterdam |
| Edge cache lifetime | tied to the link |
| Deleted data purged from all regions | within 60 seconds |
Not an exhaustive list — just the accounts we open most often.
Masters and dailies heading to a client who opens them exactly once, and rushes coming back from a shoot two timezones away.
Build artifacts handed to testers on a link that dies with the release candidate, and crash dumps gathered from machines in the field.
Document sets going out under a deadline and evidence packs coming in from a dozen counterparties who will never create an account.
Managed transfer for files that are too large, too many or too confidential for an email thread.