If you really wanted to transfer lots of data and cared about the cost, streaming or other forms of “online delivery” do not often make sense. In other words, if shipping a few hundred gigabytes of data, it’s generally faster to FedEx a hard drive than to send the files over the Internet.
A company with two offices in different cities or countries, equipped with 100 megabit Internet access connections, could transfer about 45 gigabytes in one hour – or just over one terabyte of data in 24 hours – if all the paths were working optimally all the time.
For anything larger than one terabyte, it would be faster for this company to just send the data on disks for over-night delivery. Granted, few consumers have common needs for file transfers so large.
But there is a consumer analogy, and it’s called television. As any engineer in the communications field might say there are different ways to handle any computing process. One can use local storage and local processing or substitute remote processes using communications.
“Sneaker net” approaches essentially substitute offline methods for online methods.
Broadcast television or IP multicasting are efficient ways to transfer bandwidth-intensive full motion video “files” compared to “on-demand” unicasting, for example. Netflix DVDs are another way of getting content moved back and forth between a supplier and a consumer.
The point is that it sometimes makes more sense to use a network to move files, and sometimes makes more sense to use some offline mechanism. Likewise, it sometimes makes more sense to use unicasting or broadcasting, rather than full on-demand delivery.
Using a digital video recorder with broadcasting creates something of a video on demand capability, with far less use of the communications network, for example, for high-demand content that has to be delivered at low cost.
On-demand delivery tends to work better, economically, for low-demand content.
To use an analogy, the FedEx fleet of planes, carrying nothing but hard drives, could deliver 100 times the daily volume of Internet traffic in a single day.
That’s an extreme and unworkable example, but illustrates the choices. If cost matters, there might be good reasons to use different mixes of communications and local storage, communications and offline transfer, for consumer video files.
Some 93.9 percent of mobile Internet access subscribers in the United States can buy access at 3 Mbps or faster, compared to about 93 percent of fixed network locations, an NTIA analysis suggests.
Some 34 percent of homes have access to fixed wireless networks offer access at 3 Mbps. About 86 percent of locations can buy cable high speed access at 6 Mbps or faster.
Sixty-four percent of digital subscriber line locations are able to get 6 Mbps service, and about 78.6 percent of locations have access to mobile broadband of at least 6 Mbps.
Availability begins to diverge more at speeds of 25 Mbps. Only about 7.7 percent of U.S. homes have access to DSL at that rate. But 75.5 percent of homes can buy cable modem service operating at 25 Mbps.
About 4.7 percent of homes can buy fixed wireless service at 25 Mbps.
That will change, of course. In 2002, only about 10 percent of U.S. households were buying broadband service. A decade later, virtually all Internet-using households were buying broadband access service.
Researchers at Technology Futures continue to suggest that 100 Mbps will be a common access speed for perhaps 20 percent of U.S. households by 2020. Perhaps another 20 percent will buy 50 Mbps connections while perhaps another 20 percent will buy 24 Mbps service.
The point is that even with faster access speeds on the way, there will always be economic choices that suggest other approaches, including broadcast, IP multicasting and use of local storage, will often be "better" ways to deliver big file such as TV shows and movies.
Edited by Braden Becker