Slashdot Mirror
Cornell Implementing Bandwidth Charges
Sabalon writes "Cornell University is planning on implementing a plan where if faculty, staff or students use more than 2GB of bandwidth a month, they will be charged for the additional bandwidth usage. The article mentions that last year over 100,000GB worth of files were sent from Cornell's network. I'm sure this is not the only school doing this or moving to this. I'm sure the conspiracy theory people will see this as a suggestion by Microsoft to stop students from getting those pesky Linux iso images. At least, according to the RIAA, CD sales around Cornell should now skyrocket :)" It'll be interesting to see how this plays out. Since students often have accounts on several different university machines, I suspect the more rebellious ones will be running an assortment of proxies and redirections to get around the restrictions.
Will the charges be based on MAC address? Since MAC addresses are so easy to spoof, authentication will become necessary.
Most recently-made switches can be set to only allow a single MAC address per port.. This would fix their problem with hubs as well as prevent MAC spoofing. Some can also be set to only allow the first MAC address that they see on a port and then lock out any new ones, making administration a little easier.
At my school, if you download just about anything during the day, or download anything over aboug 5 megs at full speed (about 1.5megabytes/s - its an oc3) you simply get cut off. No questions, and no getting it back.
Easily.. Our school uses a Cisco VPN solution to authenticate students accross the wireless network. Your MAC address is then attached to your student ID. I would imagine they could easily record bandwidth that way. And yes, they have Linux clients for this configuration :)
The problem is really that most p2p software doesn't make much of an attempt to take the physical network's topology into account when it creates the virtual network of peers.
Years ago, before napster took off, I described what was essentially an idea for streaming p2p (didn't call it that) to a friend who is a very smart networking specialist, and he was horrified. I think he had visions of chunks of video being passed from kansas to hong kong to iowa to france, etc. I was too lazy and not skilled enough to follow up on my idea, so I lost my place in history.
But my friend's criticism was valid then, and it's valid now, and as these services become more popular, the chickens are coming home to roost.
It seems to me that if p2p software allowed people from a specific school to look for files on each other's computers first, and to go outside of the campus only when necessary, a lot of bandwidth would be saved.
Actually, the policy at Vanderbilt is 1GB down, .5GB up per day. If you exceed those limits, you get capped to 64kbps each way. This policy seems to be effective in limiting abuse of bandwidth, but still allowing legitimate uses.
I've only heard stories of service being revoked in cases of copyright infringement.
No, but bandwidth is like sewer pipes: you NEVER try to fill it up. At 10% you start dropping packets. You go much above 25% full and service is degraded. At 50% the network is pretty much useless. I DO work in a NOC. I DO know.
Not so much as you think, else you wouldn't be quoting statistics for unswitched ethernet traffic. a T1 is a point-to-point link - you can use 95% of its bandwidth without dropping packets. Sure, you don't want to go above 50%, but that's mainly so you have room to grow.
"We returned the General to El Salvador, or maybe Guatemala, it's difficult to tell from 10,000 feet"
It sucks.
/Very./ We're split into two campuses--medical and academic (perhaps you've heard of MCV Hospitals)--and we're actually not even in the same ZIP code, we're so large.
... We have big-time pipes.
;-) (700 KB to 1+ MB from Akamai-backboned stuff like downloads from Apple.com, for example). Go VCU.
...
...
/sold,/ in the salesman sense--the university new networking hardware. I don't know what we replaced (some have told me it was Cisco hardware) but the decision has been one with terrible consequences.
/could,/ we /would have by now./ I wonder exactly what we've "upgraded" to/with, because it doesn't seem to be doing a very good job.
/your fault./ ... But no, that's not how things are.
/in writing/ for the students that abuse is against university Internet usage policy, look for abuse, track down the abusers, and actually enforce it. Abusers lose their bandwidth entirely, for example. Repeat offenders go on academic or residence hall probation.
/more/ gets abused and eaten up by the same people asking for it.
/really/ pissed off were it not confined to the residence halls. ;-)
This is a given.
But what else are universities to do when their asshat students leave crap like KaZaA on 24/7?
Here's a nice little essay for you:
Here at Virginia Commonwealth University, we have very serious bandwidth.
We have a lot--a LOT--of bandwidth in order to support the two campuses and the hospitals here. We have OC-3s and DS-3s and GTA-3s and See-Deez-Nutz
Here's our problem.
The residence halls alone are currently using 50% of the entire university's bandwidth to the outside world. Of this amount, roughly 90% of that is taken up by P2P apps like KaZaA and Gator.
I'm not making this statistic up, sadly.
The students come to us and say, "Why doesn't VCU just give us more bandwidth?"
We always reply, "Because that's what we used to do, and the only thing that happened was more file sharing."
This is simply obvious to any Slashdot reader.
Good: VCU has quarantined residence hall bandwidth. At my desk, I get great speed everywhere that hasn't been Slashdotted or Farked.
Bad: This is easily solved, but not with
Well
Welcome to state schools. Apparently, some time in the past year or two, someone (I believe Dell, as the university higher-ups are suckers for anything with Mike D.'s logo on it, regardless of if it's not the best purchasing decision for the situation) sold--and I mean
Apparently, VCU doesn't have the ability to do anything at the individual node level, like impose speed or usage caps, other than turning a port ON or OFF; the only thing we've been able to do is quarantine the res halls' subnets from from the rest of the university.
I say this because it seems logical that if we
At least, that's what I would have done, had I any control over the situation--every student gets, essentially, N amount of kilobits/sec. Something reasonable, something fair. It doesn't take a TON of bandwidth to be enough for students in the residence halls who need to check their grades, download assignments, and do research--even enough for gaming and doing light Internet file sharing. (Where inter-LAN sharing doesn't have that restriction, of course.)
The students are responsible for themselves, like the adults they technically are. Wanna trade files over KaZaA all day? Fine, it's your node allotment, use it any way you wish. Don't complain about slow Internet speeds anymore, 'cause it's clearly demonstrated that it's
Of course, I'm not stupid. That has issues associated with it, too. Another idea would be to put
No pun intended, this is all academic.
Everyone's at fault in a lot of universities, whether it's the students and a bad IT department or the students and bad adminstration. Here, the students want more to get work done and such, but when they get more, the
No, not everyone's abusing it. But as anyone on teh Intarweb knows, it's a very large percentage of the people. And even if it isn't, and the minority is hurting the majority, it's still up to BOTH the adminstration/IT people, to police their turf--and the students, to police themselves.
I'd be
-/-
Michael Watson
Apple Service Representative
Virginia Commonwealth University
http://www.vcu.edu/
Mikey-San
Karma: +Eleventy billion (mostly affected by watching Celebrity Jeopardy)
The anonymous donation was to put wireless in all the libraries, NOT all over the campus. The access points were removed from the dorms because CIT and ResLife couldn't come to an agreement over who would pay for them.
Why bother flashing the EEPROM?
ifconfig eth0 hw ether de:ea:db:ee:f0:00 is all you need. (You may need to bring the interface down first, though.) Additionally, it's not as a student couldn't wait until the target machine went into sleep mode or was shut off before spoofing its MAC address.
I agree with those posters who say Cornell's network is solely for educational purposes. As long as Cornell provides access to outside broadband providers (cable, xDSL, FSO, wireless), there should be no problem with people putting two NIC's in their machines and dual-homing them. I mean, shit. I can pop down to CompUSA and get a 100baseT PCI NIC for about $10. Bottom line: the school is obligated to provide for students' education, but not their entertainment. Another solution is for Cornell to completely get out of the business of providing connectivity to dorms and open it up to those companies providing access to MDU's (multiple dwelling units) -- and there are plenty of those companies. That way, the economics would cease to be distorted and those who use up a resource would have to pay proportionally. It's the same argument with water. I think it's silly that many apartment complexes include unmetered water useage with the rent. This distorts the allocation of this resource, as some people will wash their SUV's daily, whilst others hardly use water at all.
'He who has to break a thing to find out what it is, has left the path of wisdom.' -- Gandalf to Saruman
Could something like this turn out to be a real boon for Freenet to get a critical mass of users in one area?
fencepost
just a little off
We do have a (seemingly unautomated) system of discouraging Kaaza users. Once a student has transfered over 500MB of bandwidth within 24 hours, their speed is throttled down to the speed of a 56K modem. When I made the switch to Linux a couple months ago, I downloaded roughly 1.7GB of ISO's -- all in one day. My bandwidth was never reduced, which leads me to believe that they are either not strict about it, or that they actually look to see where the traffic is coming from, and act accordingly
I'm at the better in-state school, and I've worked with the network administration here on solutions to bandwidth problems. The way I understand it, the upstream ISP for both of our colleges will periodically (daily IIRC) send a list of the IP addresses with the highest bandwidth usage to the network administration here. These users are then placed in a sort of "penalty box" -- if it is determined that their high usage is due to not-so-nice things such as P2P, their bandwidth gets throttled back.
Here at OSU, though, they implemented a totally different solution at the beginning of this semester. Students on the ResLife network are now by default placed behind a NAT configuration. If you want a public IP, fine, but you have to register for it. Thus, if you have a public IP and your IP starts sharing illegal files and generating high bandwidth usage, they don't even have to try to figure out who you are. This has been working out nicely so far; it's much better than the old configuration, in which the severely capped ResLife network was so clogged it was hardly usable. Now, there isn't any cap, and available bandwidth is plenty.
I pledge allegiance to the flag...
of the Corporate States of America...
This is quite true. I was a member of a focus group on this idea, and they said that use of the internal campus network was consistently at about 2% of capacity. External capacity has hit 100% in the past, causing them to limit Resnet traffic to a certain percentage of capacity. They would be quite happy if file-sharing were more internal. (Happy, that is, to the extent allowed by law. Cornell complies (somewhat grudgingly) with the DMCA.)
over here, in the UK, ntl have imposed a cap on customers.
:/
Seems daft, since the reason most people get cable or DSL, or pay for this type of service, is for P2P or for things that need the bandwidth