IoT - The end of the internet

Owen DeLong owen at delong.com
Thu Aug 11 05:39:34 UTC 2022


> On Aug 10, 2022, at 15:29 , Christopher Wolff <chris at vergeinternet.com> wrote:
> 
> Hi NANOG;
> 
> I appreciate all the thoughtful replies and I apologize for vague posting when I should be sleeping.
> 
> Let me paint a little more context and hopefully this will help inform the conversation.
> 
> Use Case 1:  Augmented Reality/Virtual Reality.  It is stated that round trip latency must be <4ms with 100mbit full duplex at the cell edge to prevent nausea and dizziness while wearing goggles for a long term.  

That’s only true if you’re trying to send stereo full frame video to the goggles from a remote location. If you have intelligence on the user device and can render a lot of the stuff locally, that bandwidth requirement drops dramatically.

> Use Case 2:  A little closer to “IoT”. An autonomous vehicle under remote control requires 100 feet to stop with LTE vs 20 feet with 5G.  

An autonomous vehicle shouldn’t be taking cellular data into account for stopping distances… Onboard sensors should be able to stop the vehicle when time is critical.

> Use Case 3:  A Lidar near-miss sensor at an intersection requires 1ms from the traffic operations center.

I’m not sure I understand the meaning of this statement. Is the traffic operations center controlling the vehicles approaching the intersection? Why would the vehicles not be able to sort this out autonomously?

> I hypothesize that there is a ‘breaking point’ between safety, health, and latency and traditional IP.  

I hypothesize that if you are doing life support or life critical operations over traditional IP, you are doing something very very wrong and people will suffer dire consequences as a result.

> Will tomorrow’s applications require a re-thinking of “The Internet” and protocols that are low latency compliant?  Will we be building an infinite number of mobile edge compute boxes?  

It sounds like your idea of how tomorrow’s applications will operate will require some re-thinking. I know my Tesla, for example, when in full self-driving (beta) mode does not phone home
before it decides to hit the brakes, swerve, or take other emergency actions for example.

> If there’s an academic study describing this potential issue it would help kickstart some interesting research.

I think that the issue will usually be obviated by moving the time-critical decisions closer to the edge (or never centralizing them to begin with).

Owen

> 
> Best,
> Christopher
> 
>> On Aug 10, 2022, at 1:26 PM, Alexander Lyamin <la at qrator.net <mailto:la at qrator.net>> wrote:
>> 
>> It's not devices. It's software and what's worse protocol specifications that are implemented in this software.
>> 
>> And we still didn't get the memo in 2022. Some colleagues think that having builtin 5x Amplification in protocols freshly out just this year "is OK".
>> 
>> ....  Cyberhippies.... 
>> 
>> On Wed, Aug 10, 2022, 05:12 Ca By <cb.list6 at gmail.com <mailto:cb.list6 at gmail.com>> wrote:
>> 
>> 
>> On Tue, Aug 9, 2022 at 7:23 PM Christopher Wolff <chris at vergeinternet.com <mailto:chris at vergeinternet.com>> wrote:
>> Hi folks,
>> 
>> Has anyone proposed that the adoption of billions of IoT devices will ultimately ‘break’ the Internet?  
>> 
>> It’s not a rhetorical question I promise, just looking for a journal or other scholarly article that implies that the Internet is doomed.
>> 
>> In so much as IoT devices are ipv4 udp amplifiers
>> 
>> https://www.ndss-symposium.org/ndss2014/programme/amplification-hell-revisiting-network-protocols-ddos-abuse/ <https://www.ndss-symposium.org/ndss2014/programme/amplification-hell-revisiting-network-protocols-ddos-abuse/>
>> 
>> 
>> 
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.nanog.org/pipermail/nanog/attachments/20220810/4b21060c/attachment.html>


More information about the NANOG mailing list