Cloud // Infrastructure as a Service
Commentary
3/13/2013
10:30 AM
Keith Fowlkes
Keith Fowlkes
Commentary
Connect Directly
RSS
E-Mail
50%
50%

Can Colleges Tame The Bandwidth Monster?

Not likely. But OpenFlow and other new networking technologies provide some hope.

Imagine that you're starting college today. Your dorm room -- excuse me, residence hall suite -- has a cable TV jack and an Ethernet jack in the wall. But you didn't think to bring a TV or a device that even has an Ethernet connection. The college has wireless access, but it's far too slow to stream your regular TV programming to your tablet computer. Your roommate has a laptop that can use that Ethernet jack, but he says it's even slower than your wireless connection. You try your cell phone and you get two bars inside the room, at best. You need to download your textbooks for the semester, but more than 2,000 other students are doing the same thing at the same time. You want to call home to complain, so you have to run outside to get a cell signal because the wireless network isn't fast enough to Skype.

You begin to panic, and then get that sick feeling in the pit of your stomach. You have been (dun dun dun) disconnected from the world!

As a higher education CIO, I often try to put myself in the place of students, and this is the most dramatic scenario I've come up with so far. Unfortunately, for many college students this scenario isn't far off the mark. Welcome to college life in our digital society.

Each year, I'm faced with the continual challenge of feeding the Internet bandwidth monster. It's a strange industry when higher ed CIOs are elated about the next big thing in technology while also terrified by the effect it may have on our already congested networks and Internet connections.

[ Are printed textbooks on the way out? Read E-Textbook Pilot Puts College Books In Cloud. ]

Maryland-based State Educational Technology Directors Association recommends that a national standard for bandwidth be set for K-12: 1 Gbps per 1,000 students and faculty/staff members for Internet connections and 10 Gbps for LAN connections, all by the 2017-18 school year. If this is to be the standard for primary and secondary schools, what might the standards be for a small college or university? And what about our major research universities? 100 Gbps to the desktop? Terabit backbones? Higher? The nationwide costs could be staggering for the necessary wiring and hardware upgrades. Hang on -- I think I need to lie down a minute.

How Long Can We Feed The Monster?

Today, adequate Internet access is tied so directly to academic scholarship and research that we no longer ask ourselves whether adding more bandwidth is really worth the trouble and expense. For work or play, fat pipes are now just part of student expectations, like electric lights … and a campus coffee bar.

I've tried to keep up since the early 1990s with all sorts of network solutions. First, we threw more bandwidth at the problem. Then we made "creative use" of filtering tables, then packet shapers, now bandwidth equalizers. But we never solved the problem for long as the newest "disruptive" technology clogs our pipes.

With applications such as Skype, Netflix and Xbox, our bandwidth situation was bad enough, but now with cable and satellite content providers piling into the mix with their mobile apps, how do we keep up with demand? Even more disruption is coming with new augmented-reality applications and devices, so gear up!

Can We Tame The Monster?

If the monster is demand, my answer to the above question is, "No way." But I'm hopeful that technology innovations will keep us a step ahead of growing demand.

WAN research through the Internet2 organization lights the way to the future. Rod Wilson, senior director of external research at Ciena Corp., wrote in a 2011 blog that 100-Gbps long-haul transmission and switching have solved myriad "well-publicized problems" for providers of communications services to research institutions, "including bringing relief to saturated fiber plants suffering from nearly full links." Commodity ISPs are just starting to deploy these more-efficient, higher-speed technologies, promising higher-bandwidth services at lower cost.

So much for the wide area. But how do we attack the head of the bandwidth monster, the constantly growing need for faster wired and wireless connections inside our college campuses?

My good friend Jimmy Ray Purser, a network engineer with Cisco, says the major problem today with most campus networks isn't the amount of bandwidth but the optimization of that bandwidth. "About 40% of your traffic is overhead chatter," Purser says. "The trick is to have a good VLAN design and newer switches to capture as much benefit as possible with the latest compression algorithms. But the best features of bandwidth optimization tools are not being used in many colleges and universities today."

Purser maintains that OpenFlow could be the answer to many of our campus bandwidth problems. OpenFlow is a vendor-agnostic standard that allows network switches to be controlled by a central network operating system. With OpenFlow, packet-forwarding instructions set up in the NOS make more efficient use of processing power and paths. This technology is now being tested in switch-to-server and switch-to-cloud (cloud bursting) applications, allowing for the next level of resource optimization and customization across the entire hardware infrastructure.

Higher ed IT organizations will still continue to deal with the challenges we have come to know and hate: file-sharing software, illegal music/video downloading, gaming and streaming video services. But the solace is that OpenFlow and technologies like it will provide a leap forward in campus network bandwidth and management over the next five years.

Will costs come down? I'm pessimistic. Where there's pent-up demand, there's always a higher price to be paid, even if carrier competition increases over the coming years. The key to the future is balancing bandwidth with new technology to slow the monster and keep it from swallowing us whole.

Additional Questions

I continue to have more questions on this topic than answers, and I'd like to get your ideas. Over time, will LTE and next-generation cellular transport technologies ease the strain on campus networks? Will bandwidth provision continue to be the responsibility of the institution only, or is it fair to ask students to augment college-supplied Internet service with their own data services? How would this shift from private campus networks to public cellular networks affect security?

Please weigh in with a comment below.

Can data analysis keep students on track and improve college retention rates? Also in the premiere all-digital Analytics' Big Test issue of InformationWeek Education: Higher education is just as prone to tech-based disruption as other industries. (Free with registration.)

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Keith Fowlkes
50%
50%
Keith Fowlkes,
User Rank: Apprentice
6/11/2013 | 9:33:16 PM
re: Can Colleges Tame The Bandwidth Monster?
That would be fine if we were providing standard commercial service to students but there are many, MANY more things to deal with in higher educational networks- security, PCI/DSS compliance issues, RIAA regulations, research networks access, FERPA/HIPPA regulations and much more. I really wish it were that easy to privatize the service but it really isn't.

Thanks for the feedback!
RalphEP
50%
50%
RalphEP,
User Rank: Apprentice
4/17/2013 | 3:04:11 PM
re: Can Colleges Tame The Bandwidth Monster?
You could install a PacketViper from Viper Network Systems. It is an Intelligent Country Network Filter that would provide an extra layer of security, while also saving Bandwith.
awittmann941
50%
50%
awittmann941,
User Rank: Apprentice
3/14/2013 | 4:35:28 PM
re: Can Colleges Tame The Bandwidth Monster?
This doesn't really seem to be that hard of a problem to solve. Let's say your institution has 5000 students and you charge $15000 a semester for tuition - that's $75 million a semester. If you think it's important to provide anywhere anytime high performance internet access as a way of justifying that tuition, then figure on providing each kid with 20 mbps of access, and carve out $30/month per kid from that 75 million to do it. If you can't do it for that, Cox or Comcast can. They do it for the rest of America, they can do it for your campus too.

There's no shame in turning to a private provider, at least for the dorms and possibly for your public access wifi. For the five months of the semester, that's $150 / kid. Higher ed institutions have no excuse for not doing this right. You can even let the kid and his parents pay the provider directly. Then mom and dad can fight with junior about just what access he really needs and whether HBO is a must. When you attempt to provide it as an entitlement included in the tuition, you'll always be a bad guy.

Free enterprise is good.

Oh, and future luddite - two words: Anger management.
David F. Carr
50%
50%
David F. Carr,
User Rank: Author
3/14/2013 | 2:59:56 PM
re: Can Colleges Tame The Bandwidth Monster?
There ought to be a way to distinguish between the consumption of video lectures for a "flipped classroom" versus sucking down movies and Dr. Who reruns. However, universities might want to think twice before blocking or limiting access to entertainment content, given that the institutions who provide the best bandwidth are starting to use that capacity as a recruiting tool.
pwndecaf
50%
50%
pwndecaf,
User Rank: Apprentice
3/14/2013 | 2:23:27 PM
re: Can Colleges Tame The Bandwidth Monster?
Spoken more like a "current" Luddite. You didn't have a tv - awwww - so no one should need more than you had.
ChrisMurphy
50%
50%
ChrisMurphy,
User Rank: Author
3/13/2013 | 10:36:34 PM
re: Can Colleges Tame The Bandwidth Monster?
I had a similar first reaction, Future Luddite, that there must be some kind of metering and monitoring that can keep college bandwidth costs from soaring in order to keep Netflix revenue flowing. But you lost me with the assumption that video sharing's frivolous. A lot of our collaboration and information sharing's going to happen by video as we carry better and better video devices (smartphones) with us all the time: "Watch me do this experiment to see if I'm doing it right; let me share that engineering drawing; take a look at my presentation and see what you think." I think Mr. Fowlkes is right to keep searching for technology answers for bandwidth, because the demand will only grow.
Drew Conry-Murray
50%
50%
Drew Conry-Murray,
User Rank: Ninja
3/13/2013 | 9:41:45 PM
re: Can Colleges Tame The Bandwidth Monster?
I sympathize with Mr. Fowlkes. The kids coming onto to campus have essentially grown up online and expect ubiquitous connectivity, both for legitimate academic purposes and for entertainment. Colleges and universities are also in a difficult position because students (or their parents) are laying out outrageous sums of money, and so students expect a commesurate degree of services. I'd be interested to hear more about the use of OpenFlow and network virtualization to help improve bandwidth utilization.

Drew Conry-Murray
Editor, Network Computing
Future Luddite
50%
50%
Future Luddite,
User Rank: Apprentice
3/13/2013 | 6:02:02 PM
re: Can Colleges Tame The Bandwidth Monster?
I understand the quantitative challenges for educational institutions trying to keep up with the ever more wasteful and self indulgent uses of bandwidth that seem to be endemic in our society. However, I think you are missing the opportunity to qualitatively challenge the assumption that your job is to stay ahead of the curve of endless waste that seems to drive commercial uses of the Internet and cellular technologies rather than trying to instill a sense of social responsibility in a generation suffering mightily from "connection entitlement".

Like the dramatic reductions in costs of hardware that have driven endless cycles of bloated programming and frivolous GUIs in previous decades, the fall in costs of internet bandwidth and cellular capacity in recent years has spawned various innovations designed to clog the infrastructure of the internet with a wealth of dubious self generated (and generally self absorbed) content that costs almost nothing to produce and is not worth a great deal more than that in the grand scheme of arts and entertainment. It has also driven the fragmentation of mass distribution models for legitimate media content like music, television and movies to the point where a television or radio program that used to be broadcast efficiently to millions of households is now broadcast in duplicate millions of times through file downloads and streaming services which clog the commercial arteries of our digital infrastructure in the name of mindless convenience for viewers bent on having endless options and micromanagement control over their entertainment experience. And in the process, the "haves" of our society who can afford to "custom pipe" their entertainment in via the internet have begun to threaten the economic structures of commercially sponsored radio and television which have provided the "have nots" with cheap mass access to news and entertainment for over 90 years.

If our institutions of higher education cannot start a meaningful dialogue as to the social utility of students sharing a "cute puppy" video with 15 "friends" via smartphone that incidentally drops somebody else's 911 call from their local cell tower, I fear that nobody left in society will be able to do it. Certainly, the progress of society requires efforts like you mention to make the most efficient use of available bandwidth in a technical sense, but I think it is also time that the best and brightest of the next generation pull their heads out of Facebook and YouTube long enough to think about how our use of the internet has been evolving as a society and whether an institution committed to knowledge, be it small college or major research university, should be in an endless race with its students to support the latest and greatest dumbed down ways to use a distribution channel of great potential and purpose.

As you well know, a student can access a semester's worth of text books using a lot less bandwidth than watching a 35 second video of a fraternity belching contest, or perform hours of research using remote databases for the bandwidth required to file share one bootleg feature film. From a management standpoint, I suppose one answer would be to lock down the institutionally provided systems for legitimate academic uses, and to allow those students who could afford it to access their entertainment in whatever ways they desired in the existing wasteful and competitive capitalist system. As someone who could not afford his own TV in the 1970s, I would guess that this would be a way to provide the less advantaged students with a way to excel in comparison to their more entitled counterparts. However, I think you sell the cause of higher education short by assuming that bandwidth is only a technical problem to be solved in a seamless and invisible way to be sure that existing students don't post negative comments about your IT infrastructure that will deter future admissions. There is a social dialogue waiting to be had on this subject, and if people in academic settings are not able to start it, I have little faith that the forces who have harnessed the internet to broadcast "stupid human tricks" and the venture capitalists who support the inventors of the "next great self indulgent thing" will ever broach the subject with the leaders of our next generation.
Multicloud Infrastructure & Application Management
Multicloud Infrastructure & Application Management
Enterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - June 10, 2014
When selecting servers to support analytics, consider data center capacity, storage, and computational intensity.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Archived InformationWeek Radio
Join InformationWeek’s Lorna Garey and Mike Healey, president of Yeoman Technology Group, an engineering and research firm focused on maximizing technology investments, to discuss the right way to go digital.
Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.