Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Research shows that nearly every U.S. cell provider is doing throttling

news.northeastern.edu

Post reply on HN
Print view View original

91–100 of 180 posts

Re: Research shows that nearly every U.S. cell provider is doing throttling

#91

Post by largespoon »

ddtaylor wrote: post #33
sbr464 wrote: post #25

Years back, I operated a small ISP to serve a need until cable broadband arrived in the community. I had to monitor overall usage and throttle clients frequently with high torrent/video usage. It was also an issue with hotspots in RV Parks/campgrounds where there was a higher ratio of video usage. There are technical limitations to spectrum/bandwidth, I'm curious why this isn't mentioned more in discussions like thes…

I think the issue is that most ISPs are not transparent and their advertising comes off as double-speak. Watching their promotional materials they attempt to mislead consumers into thinking they are getting unlimited connections they are free to do whatever they want with when in reality they are getting a very specific bandwidth limit with a lot of strings attached.

The problem is that no consumer will buy internet advertised as conditionally rate limited.ISPs have no reason s to stop what works.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#92

Post by nbhuik »

_red wrote: post #85

Research shows you should use the proper name: QoS The name "Throttling" is an attempt to frame this pejoratively. All ISPs of every type engage in bandwidth shaping / QoS. Its the laws of nature that enforce this (demand always is greater than supply of every resource).

Yes, it isn't particularly hard to figure out why they are throttling. Video services especially will take up all the bandwidth when they can and then annoy people when it inevitably gets downgraded again. For most people "speed" is "does my video load" and "does this file complete fast", not "4k everywhere". This should be fairly obvious for anyone who have even shared a connection with a couple of people. And unfortunately the Internet is lacking in support of adaptive anything.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#93

Post by atonse »

_red wrote: post #85

Research shows you should use the proper name: QoS The name "Throttling" is an attempt to frame this pejoratively. All ISPs of every type engage in bandwidth shaping / QoS. Its the laws of nature that enforce this (demand always is greater than supply of every resource).

Nah, just read the article:

“There’s no evidence that any of these policies are only happening during network overload,” said Choffnes. “They’re throttling video traffic even when the network doesn’t need to. It happens 24/7, and in every region where we have tests.”

That's not QoS, that's greed.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#94

Post by belorn »

timr wrote: post #80
belorn wrote: post #52

Earlier quoted context omitted.

The article brings up that. The throttling was not done in order to ease congested network traffic. It is not a response to hotspots of high traffic, nor is it even classification of data types such as torrent and video usage. If you use Boost ISP and watching sports on NBCSport it will work fine but watching the same match later on youtube is not. If you sit at CSpire and watch a paid movie on Youtube Red you will g…

"The article is not about congestion control but rather marketing choices where one streaming service is favored over the other within the captured market of an ISP customers." I wonder if that's true, or this is simply a result of the peering agreements that content providers have engaged in since forever. For those who don't know: if Netflix is colocated in the same datacenter(s) as CSPire, it's possible to "throw…

Could there be other causes than differentiation to explain the data collect? That would be an excellent questioning to ask the team at University of Massachusetts if they made a dissertation or research paper, since all the article says is that they used a established peer-reviewed technique that assumingly has gone through such questioning. It would be interesting to know the statistical probability that collocated servers would exclusively explain the data.

> now that I've read TFA, it's pretty clear that they're not talking about favoring providers, but rather, throttling entire classes of traffic.

I must strongly disagree there. What kind of different class of traffic would give throttled traffic to different forms of on-demand video stream services? NBCSports, Netflix, YouTube Premium and Amazon prime. I don't see it so please name the four different classes of traffic those sites represent that a congestion system would use as identifiers.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#95

Post by rayiner »

joncrocks wrote: post #82
nbhuik wrote: post #79

Earlier quoted context omitted.

That is just how the Internet works. Everyone is selling and buying access to networks. Whether directly or in the form of agreements. And most people are of course trying to get more for less. One could certainly wish that things would be different, but it isn't like they are doing something especially nefarious.

Imagine that you have a delivery service. You pay your local delivery company $X per month and they promise that they'll deliver you 1 package a day from their local depot. But it turns out that although you're ordering a package a day from Amazon, you're not receiving a package a day. You get your stuff, but it's delayed and backed-up somewhere. And so as an experiment you sign up for a service that will receive a p…

That’s just not how the Internet works and it’s never how it worked. This old article from Cisco (based on a 1999 book) explains that there can be various kind of business arrangements in an interconnected network: https://www.cisco.com/c/en/us/about/press/internet-protocol-.... The article likens the situation to airlines, where multiple transit providers might be involved and where there may be complex fee sharing arrangements between all the providers.

Note that this book predated the rise of streaming video providers by more than a decade. Back then, nobody thought that delivery of content over the Internet was this simple one-sided transaction. It’s a new view of the Internet that has arisen because it suits the interests of these very lucrative streaming businesses.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#96

Post by rayiner »

chii wrote: post #66
pathseeker wrote: post #57

Earlier quoted context omitted.

What you're asking for already exists. They are dedicated business circuits. I get where you're coming from, but there is a reason that the consumer advertisements always say "up to X mbps". If ISPs didn't oversell, consumer Internet would either be unbearably slow or obscenely expensive. A guaranteed 1 gig circuit is on the order of thousands USD month. An oversubscribed consumer fiber provider like Google Fiber is…

> Are you okay with gyms selling more unlimited memberships than the number of machines they have? Yes, but with a gym, you can switch if you keep seeing overcrowdedness. Not so with an ISP that has a monopoly.

First, that’s not relevant to this particular thread, since there are several major wireless ISPs. Second, it’s besides the point. All gyms oversubscribe because that’s the only sensible model for consumer access. If you didn’t allow oversubscription, Google could sell only a 75/35 service (each GPON has 2.4G down /1.2G up shared by 16-32 users). Would that be better?

Re: Research shows that nearly every U.S. cell provider is doing throttling

#97

Post by berbec »

sbr464 wrote: post #58
DougBTX wrote: post #53

Earlier quoted context omitted.

Personally I’ve had a great experience with 4G LTE, but it is going to depend on lots of factors, including the ISP.

Curious, regarding monthly download caps, has anyone found a solution to using a 4G LTE connection for frequent Netflix consumption? It currently burns through too many gigs to be a cable internet replacement. I was looking for a solution for doing longer term domestic (US) travel.

Some providers, T-Mobile I know, offer to bypass the data cap for certain services. The drop the quality to 480p, but that's not a huge deal on my 5" screen.

Re: Research shows that nearly every U.S. cell provider is doing throttling

#98

Post by andrewla »

I'm not entirely sure I understand what these graphs are saying, because they're somewhat confusingly presented.

Specifically, I see no evidence of differentiation. The data looks so sparse that I'm not even sure I understand what the signal is supposed to be. In the first chart, does an absence of a number mean that no throttling was detected, or does it mean that the combination was simply not tested?

I assume the latter, and in that case I don't really see anything interesting, especially since we're not given N or a standard deviation.

ATT, for example, shows 1.5, 1.4, 1.4, NA, NA for its data points, which means that it doesn't appear to differentiate between NBCSports/Netflix/Youtube. The other data points show similar patterns.

Am I misreading this?

Re: Research shows that nearly every U.S. cell provider is doing throttling

#99

Post by rayiner »

close04 wrote: post #72
mehrdadn wrote: post #67

Earlier quoted context omitted.

> You don't have to guarantee a minimum speed but you have to advertise the relevant parameters that you can reach at least a certain percentage of the time. But this still suffers from exactly the same problem, just in a statistical way. i.e., what if someone's frequently in a place that has poor signal? How are you going to predict what percentage of the time they'll be in such a situation?

Then you compensate that person for not being able to provide the service. Just as you'd expect in any other case when a service is not delivered. And if we're being honest, that's not actually the reason most of us get throttled. Most of the time it's not even network management, it's just a money grab. I don't intend to go into the technicalities because they're irrelevant. The point is they should offer a quality…

Wireless ISPs don’t advertise any particular speed level. They sell “best effort” service and advertise it that way. What you’re asking for is an SLA (presumably for consumer level prices).

Re: Research shows that nearly every U.S. cell provider is doing throttling

#100

Post by davidkuhta »

andrewla wrote: post #98

I'm not entirely sure I understand what these graphs are saying, because they're somewhat confusingly presented. Specifically, I see no evidence of differentiation. The data looks so sparse that I'm not even sure I understand what the signal is supposed to be. In the first chart, does an absence of a number mean that no throttling was detected, or does it mean that the combination was simply not tested? I assume the…

Agreed, I found the visualizations confusing.

Also, is there significance in the separation into two 'blocs' of carriers? I thought it was 'US (ATT-ViaSatInc)' and then 'global (du-iWireless)', but O2UK was in the former set and cricket seems to be without a home.

Post reply on HN