5 ms·

I might be all for it if it frees up some bandwidth for the rest of us. I travel full-time and quite a few places lately the Verizon tower is obviously at capa

I might be all for it if it frees up some bandwidth for the rest of us. I travel full-time and quite a few places lately the Verizon tower is obviously at capacity. (Strong signal.. slow speeds)

MichaelGG 9y ago
Limiting Netflix is a poor way to approach that; they should instead be limiting users on the tower, dividing up the bandwidth there, regardless of what you're accessing. If capping flows at 10M helps, then they should apply it to everyone.
- tedunangst 9y ago
  It's a little more complicated than that. If I click a link to an article on medium, I want 100Mbps to download it, but then I'm going to spend some time reading it, freeing bandwidth for others. There are all sorts of token bucket algorithms one can use, but like I said, a little more complicated. Identifying popular streaming services and ratelimiting them might be easier than tracking traffic usage per user per minute. And sometimes it's ok to let users download at peak speeds for more than a web page or two at a time. Maybe I download a new ubuntu ISO, which is a couple gigs and takes a while. I don't do that everyday, so the impact is minimal. It's not just that streaming video is high bandwidth, it's also very popular, and frequent, and long duration. Ideally, we might offer users a baseline of 10Mbps all the time, and some quota of 100GB at 100Mbps regardless of source, and then people can choose what they want to download fast, but I think this would be a UX disaster.
  - ant6n 9y ago
    I don't get how 10mb/s is so much more complicated than 100mb/10s.
    - tedunangst 9y ago
      In the particular case of something like Netflix, starting an HD stream then stalling and re buffering an SD stream might be a shittier user experience. How much data does Netflix download to measure bandwidth and for how long?
      - ant6n 9y ago
        I assume they load a large buffer, adjusting bit rate to keep it full. But I don't see the topical connection with throttling people to use a certain amount of data each 10s.
  - wmf 9y ago
    This is a key issue. Do we want slightly more complex (per-customer fair queueing) equipment that is demonstrably fair or do we want fast and cheap kludges that are prone to abuse but trust us we swear they won't be used for evil?
  - shuntress 9y ago
    If I'm going to watch a show on Netflix, I want 100Mps to download it, but then I'm going to spend some time watching it, freeing bandwidth for others.
PissedWhovian 9y ago
Just curious - Why do you assume that's due to congestion and not to the same type of intentional throttling being discussed here?
- timmaah 9y ago
  The speed will decrease during periods you'd think there would be heavy usage (mid morning/late afternoon, even slower in early evening). Then speed up dramatically in the late evening when most have gone to bed.
  - dboreham 9y ago
    fwiw I see the same congestion-related throughput pattern on my copper cableco connection (fast.com reports much lower numbers in the evening and on the weekend vs early in the morning on a weekday). Typically I can still get full speed from my own servers though, implying the congestion is specific to the peering between my ISP and Netflix CDN nodes.

I might be all for it if it frees up some bandwidth for the rest of us. I travel full-time and quite a few places lately the Verizon tower is obviously at capacity. (Strong signal.. slow speeds)

MichaelGG 9y ago

Limiting Netflix is a poor way to approach that; they should instead be limiting users on the tower, dividing up the bandwidth there, regardless of what you're accessing. If capping flows at 10M helps, then they should apply it to everyone.

tedunangst 9y ago
It's a little more complicated than that. If I click a link to an article on medium, I want 100Mbps to download it, but then I'm going to spend some time reading it, freeing bandwidth for others. There are all sorts of token bucket algorithms one can use, but like I said, a little more complicated. Identifying popular streaming services and ratelimiting them might be easier than tracking traffic usage per user per minute. And sometimes it's ok to let users download at peak speeds for more than a web page or two at a time. Maybe I download a new ubuntu ISO, which is a couple gigs and takes a while. I don't do that everyday, so the impact is minimal. It's not just that streaming video is high bandwidth, it's also very popular, and frequent, and long duration. Ideally, we might offer users a baseline of 10Mbps all the time, and some quota of 100GB at 100Mbps regardless of source, and then people can choose what they want to download fast, but I think this would be a UX disaster.
- ant6n 9y ago
  I don't get how 10mb/s is so much more complicated than 100mb/10s.
  - tedunangst 9y ago
    In the particular case of something like Netflix, starting an HD stream then stalling and re buffering an SD stream might be a shittier user experience. How much data does Netflix download to measure bandwidth and for how long?
    - ant6n 9y ago
      I assume they load a large buffer, adjusting bit rate to keep it full. But I don't see the topical connection with throttling people to use a certain amount of data each 10s.
- wmf 9y ago
  This is a key issue. Do we want slightly more complex (per-customer fair queueing) equipment that is demonstrably fair or do we want fast and cheap kludges that are prone to abuse but trust us we swear they won't be used for evil?
- shuntress 9y ago
  If I'm going to watch a show on Netflix, I want 100Mps to download it, but then I'm going to spend some time watching it, freeing bandwidth for others.

PissedWhovian 9y ago

Just curious - Why do you assume that's due to congestion and not to the same type of intentional throttling being discussed here?

timmaah 9y ago
The speed will decrease during periods you'd think there would be heavy usage (mid morning/late afternoon, even slower in early evening). Then speed up dramatically in the late evening when most have gone to bed.
- dboreham 9y ago
  fwiw I see the same congestion-related throughput pattern on my copper cableco connection (fast.com reports much lower numbers in the evening and on the weekend vs early in the morning on a weekday). Typically I can still get full speed from my own servers though, implying the congestion is specific to the peering between my ISP and Netflix CDN nodes.