Hacker News Reader

Comment by AnthonyMouse | Hacker News Reader

> Bandwidth Chicken & Egg: in order to get the unit economics around bandwidth to offer competitive pricing at acceptable margins you need to have scale, but in order to get scale from paying users you need competitive pricing. Free customers early on helped us solve this chicken & egg problem.

I'm not really sure how this works.

Suppose you have paying customers and for that you need X amount of bandwidth. If you add a bunch of free customers then you need X + Y bandwidth. But the price of X + Y is never going to be lower than the price of X, is it? So even if the unit cost is lower, the total cost is still higher and you haven't produced any additional revenue in exchange, so how can this produce any net profit?

View on HN

If you send 10Gbit/s to an ISP you have to pay for transit to reach it. But if you send 100Gbit/s+ the ISP suddenly is willing to not only peer for free with you but may even host the servers for you in their data center for free. [0][1][2] So yes being bigger can absolutely save you costs.

[0]: https://www.cloudflare.com/partners/peering-portal/

[1]: https://openconnect.netflix.com/en/

[2]: https://support.google.com/interconnect/answer/9058809?hl=en

Don't forget about the Bandwidth Alliance, which is agreements for free or cheap egress between peers.

https://www.cloudflare.com/bandwidth-alliance/

Can't you just send random generated packets. Or by requesting content from other hosting provider with free or cheap egress. Or sending to another hosting provider.

Sending random packets at another AS just results in your traffic being blocked. The network engineers running these systems are smart, and the community is surprisingly small.

The thing with ISPs is the small guys are more likely to have to pay, and the smaller you are the more likely you are to pay more.

If you are a Tier 1 ISP, everyone is willing to pay you to carry their traffic and other Tier 1s just make peering agreements with you.

If you're johnscheapvps.com, you're likely to pay all your upstream ISPs for your traffic. If you're GCP or, say, digitalocean.com, everyone would love to be paying you to get faster access to all the sites hosted on your platform (and because paying you is probably going to be cheaper than their regular upstream)

Imagine you're an ISP. If your customer has slow bandwidth to some random website, they will blame the website. If they have a slow connection to YouTube, they will blame you.

So YouTube gets more favorable terms on transit bandwidth than the random site does.

it may be, especially if the ISP in question just does direct peering with you, your unit cost can drop to ~ $0/MB, and you stop paying Cogent/Verizion/HE unit cost for facilitating the connection from you to the ISP.

Works for the ISP too, one off cost for them to drop there side of the bill down

The point is that that you get your paid offering down to a lower price point because you have the volume to get the cheaper peering deals. Because your paid offering is cheap you get even more volume from paying customers which offsets the loss you made.

Reply on news.ycombinator.com