Current Information:

Although NTS's restrictions on residence hall bandwidth may be understandable (i.e. to prevent users from running dedicated servers or sharing files with programs such as Morpheus; of which are violations of NTS policy), our dissatisfaction is the fact that NTS decided not to notify residents before said restriction was implemented. As a result it has left many of our fellow students questioning and griping about the poor access speeds to utilize services that do not violate NTS policy, such as email and web browsing.

What we know is that a device referred to as "The Packeteer" was installed sometime between Feb 8th and Feb 10th. With NTS's configuration it effectively renders maximum bandwidth per computer (in the residence halls) to 1.2 Mb/sec. This has apparently been changed from the 10 Mb/sec max per computer that was installed in all of the residence halls. ResNet's website states that each etherjack in the residence halls runs at 10BaseT (10 Mb/sec). This change has severly impeded all internet related tasks for users in the residence halls, from surfing the web, to game playing. (See notes after editorial letter).

Also, stated in the ADCS email links below is that NTS is currently revising service policies. A little while back, NTS sent out a policy change form telling residence hall tenants that service policy had changed and that certain file-sharing programs were no longer allowed to be used. This current incident is similar, except the emails from NTS to ADCS show that the bandwidth "shaping" (due to increased internet usage) as they call it look like it may be a permanent impedence to internet traffic from the dorms. The email also indicates that NTS is still experimenting with The Packeteer technology, which means we could see network perfomance increases, but given the circumstances we may as well expect more hinderances, at least for residence hall tenants (for instance the email says that if someone is downloading lots of files with a file-sharing program it'll hurt everyone).

Take a look at the information we've compiled on the PacketShaper. Not being ultraparanoid, it would be nice to know exactly what is being done with this technology, and in what ways it's going to affect us in the residence halls.

To quote OIT's Network Management Page:
Goal:
No one application will affect the total network usage by consuming an inordinate amount of the University internet/intranet network bandwidth. OIT will intercede in normal traffic routing and flow as little as possible. It is not the goal of OIT to inhibit general use of the internet or campus networks.
Implementation:
OIT will set time-dependent rate limits, when needed, on any application or protocol which exceed established parameters. In order to set differentiating guidelines, we have classified applications into two separate categories. One (Primary) are those applications which have general consensus as having merit and purpose to the core mission of the University of Minnesota (teaching, education and research) and Secondary applications which may prove to have significant bandwidth requirements and as such impede core mission applications.
Primary application include:
* HTTP (web browser)
* SMTP,
* POP,
* IMAP (e-mail)
* DNS (address information)
* Other applications that use less than 7% of total internal or external network capacity.
Secondary bandwidth priorities include:
Applications not listed above that use more than 10% (or an updated percentage determined by the CIO) of total internal or external network capacity with provisions for specific exceptions (e.g. research requirements) that may be made by approval of the CIO.
Consultation and Notification:
When intervention measures are necessary, every effort will be made to work with campus populations who utilize the said application(s) to arrive at mutually agreeable methods of controlling or minimizing campus resources impact. When rate limiting methods are determined necessary, these will be put into place and approved by the CIO, who will then notify the appropriate faculty governance committees.

According to Consultation and Notification, we should have been worked with to solve the problems with bandwidth hogging applications but all we received was a memo telling us that we need to stop using the programs, with instructions on how to uninstall them. That doesn't seem like working with anyone, but rather telling them what to do.

Some say that our internet access in the residence halls is not a right, but a privilege. To quote my friend EC:

We pay big bucks for this connection (by my justification around $100/mo). It's a service I *pay* for, and they damn well better provide it in accordance with the terms I agreed upon.
...
I'm also very concerned that NTS has chosen to destabilize our network by introducing a product that appears to be, at best, beta-quality. The traffic it lets through is not a smooth, consistent stream, and it seems to add a noticeable amount of latency to our connections. As I speak, I'm downloading XFree from a sourceforge.net server at a whopping 30KB/sec, where I used to get 400+ k/sec. Now I'm not saying NTS should provide that speed, but at least cablemodem comparable speed would be helpful. As it is, I'm not even getting standard Qwest DSL type speeds. VERY bad. I guess it's time to complain to NTS.


It has been pretty obvious to those who live in the residence halls that many use file-sharing applications, but the resulting impact of the file-sharing programs has been the implementation of rate-limiting of the residence halls and bandwidth shaping of the University's network as a whole via Packeteer's PacketShaper. The question raised here is, was anyone ever going to tell us that this was going to be done and what the changes were?