Internet Traffic Still Growing Quickly
linuxscrub writes "I guess the previous articles about internet traffic doubling/[time period] being wrong were wrong? A new IDC report states that internet traffic will nearly double annually until 2007. They even use /.'s favorite unit of capacity/storage, the LOC. They predict that internet traffic will be 64,000 LOC/day! Wow, 64000 LOC, that sure sounds impressive!!"
How much is 1 loc in gigabytes?
:)
And in the article they talk about petabits. Im confused
I fought the corporate America, and the corporate America bought the law.
None of them are inspiring. I was thinking maybe I could run for election next time with just one promise, "I will work to lay the backbone for a fiber optics network to eventually reach into every house in our electorate!". Then we'll see the amount of data sent over the internet more than doubled each time :)
Of course the traffic may not be to chase after obscure documents, but simply more larger files, more peer-to-peer, more p0rn, etc...
I wonder if the traffic can be correlated back to the actual number of "transactions" that are being done on the Internet? Like when I visit a website, a lot of the traffic (large banners, pop-up, etc) aren't really what I am doing or after.
Is this simply a bandwith increase or are we talking about more real transactions? Probablly a bit of both...
D.O.U.O.S.V.A.V.V.M.
5.175 petabits is about 1 bit per square centimeter of earth.
Johan Veenstra
If the Library of Congress is 10 terabytes that's less than 300 lifetimes' worth. (Which 300 people should be included?)
Another useful measure is the EB, or Encyclopaedia Britannica, which is about 200Mb. So one LoC = 50,000 EBs = 300 lifetimes.
I wonder what percentage of that astounding amount of traffic was actually created (written, composed etc.) by the sender.
These days, most people know that "multimedia" and "software demos" make up a large chunk of internet traffic. Most of this is copyrighted by someone else.
If only there were a practical and legal way to store this information in a central depository. With widespread multicasting, the sheer amount of internet traffic would be greatly reduced.
"Smoking helps you lose weight - one lung at a time" -- A. E. Neumann
Have a look at the traffic statistics of the public peering points in Europe:
LINX - London - 25Gbit/s
AMSIX - Amsterdam - 11Gbit/s
DECIX - Frankfurt - 10Gbit/s
If you look at it most of them double traffic even faster than in 12 month. I think it's closer to 9 month.
--
Andre
180 petabits per day? What kind of measurement is that? Where was it measured? How was it measured? Who was included? Were bits counted twice?
Just to give you an idea, I work for a large IP carrier, and we peak around oh, 200Gbps aggregate traffic entering the network. Gigabits/second is a good measurement of traffic, as is total gigabytes/terabytes... but to use the term petabit, implies they're using bandwidth, not data, and that asks where that was measured and how? There's not a lot of 200Gbps networks in the world.
Current data:
180 petabits/day = 22.5 petabytes/day = 273 gigabytes/sec.
Presuming 250 million people using the Internet, that's 1118 bytes/sec for each person, or 92 MB/day. Are you doing your part?
2007 prediction:
5,175 petabits/day = 650 petabytes/day = 7.66 terabytes/sec.
Presuming 1 billion people using the Internet, that's 7,850 bytes/sec per person, or 647 MB/day. An average of one CD per day per person.
Has anyone checked the past performance of the predictors? Does IDC sell a research report that says "we were right 70% of the time for the last 5 years" that costs $4500 as well? Or maybe they could sell a report "Dot Bomb, we knew it was coming" for $4500.
I guess someone is buying these, what I want to know is how many "*BSD is dying" posts make up a LOC and are we there yet?