In an interview over on ZDNet published on Friday, VeriSign’s CTO Ken Silva made another of those crazy internet growth predictions: in ten years we will need to deal with internet traffic 1000 times as large as today. Actually, it’s not crazy, it’s simply mathematics and exponential growth with a bit of optimism thrown in and then bundled for shock value. The mathematically-oriented amongst us quickly realize that in order for the internet to grow by a factor of 1000 over a decade, it need only double each year since 210=1024.
Of course, VeriSign wasn’t talking about transport capacity but rather about server capacity, because that’s what VeriSign needs to scale. They’re doing about 60B queries daily, and are planning to find a way to handling more like 4 quadrillion. Honestly though, computational power seems to scale much more effortlessly over the years than bandwidth and with the scale-improving cloud computing movement, I don’t really think VeriSign is going to have that big a problem. The pipes though, a jump by three orders of magnitude in traffic will make mincemeat of 100G’s mere tenfold advance.
Now, internet traffic growth hasn’t been doubling annually, it’s been more like 50-75% on a sustained basis. And that rate doesn’t get us near a factor of 1000, but instead just 50-250. Yet with the bits from video in higher and higher definition plus the rising ability to burn bandwidth not just from one’s desk or couch but from practically via the iPhone and all that will follow it, it is probably quite prudent to be ready for traffic growth of 100% annually over the next decade. From the point of view of fiber operators, it would be a really good thing also – assuming prices don’t fall by 50% to match.
Let’s hope he’s right about the 1000x, and that they get started on it quickly. Or maybe we can get up to just 150% growth per year, that would mean an internet 10,000 times the size of today’s when 2020 rolls around. Exponentials can be fun.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Internet Traffic
Join Our LinkedIn Group
I’m with you Rob, right up to the point where you state “assuming prices don’t fall by 50% to match”, which is where I would ask:
What are your assumptions in this regard?
Shall price points be frozen in place or fall only slightly, keyed to 2010 market conditions, until the year 2020?
Unit pricing will continue to fall, IMO, tracking gains in optical and chip-set bit-efficiencies (ok, per “Moore’s Law”), lagging perhaps only slightly as per the usual arbitrage that takes place on the edges of market activity. Your thoughts? Anyone?
Actually, I’m not making any assumptions, just hoping that pricing pressure lags traffic growth so the segment remains somewhat healthy. But yes, unit pricing has to fall continuously, it’s the nature of the beast. The balance between the two is key…
I have to think that with the way this current administration is passing cyber terrorism legislation that the internet is not going to increase in size at all. The more regulation, the harder it is for it to grow!