[Bloat] where home 5G can go south

[Bloat] where home 5G can go south

[Dave Taht]

This had some details as to the things that could go wrong from an
initial happy install of t-mobile, to something terrible.

The author switched to verizon, but what guarantees of continued
reliability does one have?

https://blog.networkprofile.org/redundant-wan-ditching-t-mobile-5g-for-verizon-5g/

(both services had horrible bufferbloat)


-- 
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
Dave Täht CEO, TekLibre, LLC

Re: [Bloat] where home 5G can go south

[jf]

We’ve observed growing variability on some TMHI setups from our fleet, and it seems there is a correlation to usage growth on a single tower. Seems neighbors talk to hear other after all ;-)

And yes, horrible bufferbloat on these variable capacity links.

> The author switched to verizon, but what guarantees of continued
> reliability does one have?

It seems none ATM, as it really depends on user density vs tower capacity. Woe to those that share a tower with a busy highway, ‘rush hour’ likely means low capacity and even higher latencies.

Cheers,

Jonathan Foulkes


> On Oct 27, 2022, at 11:37 PM, Dave Taht via Bloat <bloat@lists.bufferbloat.net> wrote:
> 
> This had some details as to the things that could go wrong from an
> initial happy install of t-mobile, to something terrible.
> 
> The author switched to verizon, but what guarantees of continued
> reliability does one have?
> 
> https://blog.networkprofile.org/redundant-wan-ditching-t-mobile-5g-for-verizon-5g/
> 
> (both services had horrible bufferbloat)
> 
> 
> -- 
> This song goes out to all the folk that thought Stadia would work:
> https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
> Dave Täht CEO, TekLibre, LLC
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] where home 5G can go south

[Erik Auerswald]

Hi,

On 28.10.22 23:10, jf--- via Bloat wrote:
>> On Oct 27, 2022, at 11:37 PM, Dave Taht via Bloat <bloat@lists.bufferbloat.net> wrote:
>>
>> https://blog.networkprofile.org/redundant-wan-ditching-t-mobile-5g-for-verizon-5g/
>>
>> This had some details as to the things that could go wrong from an
>> initial happy install of t-mobile, to something terrible.
>>
> We’ve observed growing variability on some TMHI setups from our fleet, and it seems there is a correlation to usage growth on a single tower. Seems neighbors talk to hear other after all ;-)
> 
> And yes, horrible bufferbloat on these variable capacity links.
> 
>> The author switched to verizon, but what guarantees of continued
>> reliability does one have?
> 
> It seems none ATM, as it really depends on user density vs tower capacity. Woe to those that share a tower with a busy highway, ‘rush hour’ likely means low capacity and even higher latencies.

As another anecdote:

In the German I city where I live, Internet access via 4G usually
works well, even though there are noticeable bufferbloat effects
for those who know how to look.

But when there is a significantly higher number of mobile users,
e.g., because of a soccer game or some other event with large
attendance, latency goes up and reaches several seconds with just
light network usage.

Thanks,
Erik

Re: [Bloat] where home 5G can go south

[Sebastian Moeller]

Hi Erik,

this fits with the suspicion that the biggest different between 4G (LTE) and 5G in regards to low latency is the fact that with 5G it becomes part of the marketing effort.
IMHO the fact that 3GPPP has apparently bet on L4S as part of its low-latency effort should make it clear that this is far from solid engineering...

Regards
	Sebastian

> 
> Hi,
> 
> On 28.10.22 23:10, jf--- via Bloat wrote:
>>> On Oct 27, 2022, at 11:37 PM, Dave Taht via Bloat <bloat@lists.bufferbloat.net> wrote:
>>> 
>>> https://blog.networkprofile.org/redundant-wan-ditching-t-mobile-5g-for-verizon-5g/
>>> 
>>> This had some details as to the things that could go wrong from an
>>> initial happy install of t-mobile, to something terrible.
>>> 
>> We’ve observed growing variability on some TMHI setups from our fleet, and it seems there is a correlation to usage growth on a single tower. Seems neighbors talk to hear other after all ;-)
>> And yes, horrible bufferbloat on these variable capacity links.
>>> The author switched to verizon, but what guarantees of continued
>>> reliability does one have?
>> It seems none ATM, as it really depends on user density vs tower capacity. Woe to those that share a tower with a busy highway, ‘rush hour’ likely means low capacity and even higher latencies.
> 
> As another anecdote:
> 
> In the German I city where I live, Internet access via 4G usually
> works well, even though there are noticeable bufferbloat effects
> for those who know how to look.
> 
> But when there is a significantly higher number of mobile users,
> e.g., because of a soccer game or some other event with large
> attendance, latency goes up and reaches several seconds with just
> light network usage.
> 
> Thanks,
> Erik
> 
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat