10+ BT trackers using CF likely free tier. Low-end EACH 300-500 MILLION HTTP requests/24hr uncached, some exceed a billion reqs/day (using statistics from other open trackers) 20TB-40TB daily, ~1-3 Gbps sustained.
That's a crap load of transfer and compute to process those tiny network connections. As someone that's run a Tunnel on a normal site in the millions, the daemon uses a sizeable amount of CPU and I can't see any reason why it's not the same on the other end.
Whatever source CF has going on, kudos to them and their engineering team.
What I was talking about: https://newtrackon.com/
10+ BT trackers using CF likely free tier. Low-end EACH 300-500 MILLION HTTP requests/24hr uncached, some exceed a billion reqs/day (using statistics from other open trackers) 20TB-40TB daily, ~1-3 Gbps sustained.
That's a crap load of transfer and compute to process those tiny network connections. As someone that's run a Tunnel on a normal site in the millions, the daemon uses a sizeable amount of CPU and I can't see any reason why it's not the same on the other end.
Whatever source CF has going on, kudos to them and their engineering team.