On Jun 20, 2013, at 9:10 PM, "Aaron C. de Bruyn" <aa...@heyaaron.com> wrote:
> Why is there a variable charge for bandwidth anyways? > > In a very simplistic setup, if I have a router that costs $X and I run a $5 > CAT6 cable to someone elses router which cost them $Y, plus a bit of > maintenance time to set up the connections, tweak ACLs, etc... > > So now there's an interconnect between two providers at 1 gigabit, and the > only issue I see is the routers needing to be replaced within Z years when > it dies or when it needs to handle a 10 gigabit connection. Many things aren't as obvious as you state above. Take for example routing table growth. There's going to be a big boom in selling routers (or turning off full routes) when folks devices melt at 512k routes in the coming years. Operating a router takes a lot of things, including power, space, people to rack it, swap failing or failed hardware, OPEX to the vendor to cover support contract (assuming you have one), fiber cleaning kits, new patch cables, optics, etc. These costs are variable per city and location as space/power can be different. This doesn't include telecom costs, which may be up/down depending on if you are using leased/dark/IRU or other services. Building fiber, data centers, can be quite capital expensive. Fiber, expect 50-100k per mile (for example). It can be even more depending on the market and situation. Much of that cost is in the labor to the technicians as well as local permits as opposed to what the fiber actually costs. Many people have fiber they built 10 years ago, or even older. Folks like AT&T have been breathing life into their copper plant that was built over the past 100 years. Having that existing right-of-way makes permit costs lower, or allows you to get a blanket permit for entire cities/counties in cases. Some cable company has a presentation out there (maybe it was at a cable labs conference, or otherwise) I saw about average breaks per year. This costs splicing crews that you either have to pay to be on call or outsource to a contract company for emergency restoration. http://www.southern-telecom.com/AFL%20Reliability.pdf has some details about these. > So it seems I should be able to say "Here's a 1 gigabit connection. It > will cost $Q over Z years or you can pay $Q/Z yearly", etc... > > And wouldn't the costs go down if I had a bunch of dialup/DSL/cable/fiber > users as they are paying to lower the costs of interconnects so they get > content with less latency and fewer bottlenecks? There was a presentation by Vijay about the costs of customer support. Many states have minimum wages higher than the federal minimum wage, but even that being said, you need to pay someone, train them, give them a computer, manager, phone and other guidance to provide support for billing, customer retention and sales. I recall Vijay saying that if a customer phoned for support it wiped out the entire profit from the customer for the lifetime of them being a customer. That may not still be the case, but there are costs each time you provide a staff person to answer that phone. Sometimes it's due to outage, sometimes it's PBKAC, sometimes you don't know and have to further research the issue. Your overhead costs may be much higher due to the type of other costs you bear (pension, union contracts, etc..) vs a competitor that doesn't have that same structure. This is often seen in the airline industry. I for one would like to see more competition in the last mile in the US, but I think the only people that will do it will be folks like sonic.net, google and other smaller independent telcos. Take someone like Allband Communications in Michigan. They brought POTS service (just recently) to locations that Verizon/AT&T were unwilling to build. The person who wanted the phone service ended up having to start a telco to get POTS service there. They just went triple-play since it was the same cost to trench fiber as to put in the copper. - Jared
