5 ms·

I would understand if it was a full GNSS outage, must most "GPS" equipment nowadays is multi-constellation, using as much as four (GPS, Galileo, Glonass, BeiDou

by brohee 7y ago

I would understand if it was a full GNSS outage, must most "GPS" equipment nowadays is multi-constellation, using as much as four (GPS, Galileo, Glonass, BeiDou). The loss of one constellation should go undetected by most end users.

zaarn 7y ago
What disaster would knock out only one constellation out of four (perfectly one) and also go unnoticed to users?
- brohee 7y ago
  Politically driven shutdown. But the point I just made makes it unlikely, it's not the weapon it once was.
- jerf 7y ago
  The phrase that should strike fear into the heart of any engineer who has to work at scale (any kind of scale, not just software): "Correlated failures". A software bug that affects only one of them. The political possibility mentioned elsewhere. Some computer virus never "intended" for a satellite manages to get into one of them for some reason, most likely a very bizarre one that nobody would have ever guessed could happen but still somehow did. Some hardware component that turns out to fail all at the same time (more or less), used by all of a particular constellation. Or most likely of all, something I can't even think of to put here as an example, but still it somehow turns out that a particular failure ends up having as a side effect greatly increasing the chance of failures in the rest of the system despite all the efforts to prevent that. (The "ones we can't even think of" end up with elevated probabilities precisely because we don't think to prevent them and thus put no engineering into preventing them.) All of these outcomes are low probability, sure, but they are all vastly higher probability than sufficient simultaneous independent failures to take down the system. It's why despite all the incredible software engineering that goes into Google and AWS, they still go down sometimes. Almost no conceivable set of independent events can take those things down, but correlated failures still can, do, and will continue to do so.
- Piskvorrr 7y ago
  On the contrary, I have a hard time imagining an event that would knock all of them out and was not a relatively minor side effect of a far bigger problem, e.g. global thermonuclear war or a hypermassive solar eruption.
michaelt 7y ago
In theory - but if there's one thing leap seconds and GPS rollovers and suchlike have taught me, it's that once-every-few-years situations aren't always robustly supported even if they're entirely foreseeable.
myself248 7y ago
Maybe most equipment purchased in 2019, from top-tier vendors, with all the software options unlocked. But that's several steps away from what's actually deployed in the world right now. A tremendous amount of the telecommunications network uses BITS sync equipment deployed in the late 90s, because it was extraordinarily well built, extremely reliable, and supported over the long term. There's nothing wrong with it, but it's single-constellation. Ditto the power generation and distribution industry, mostly deployed in the early 2000s. ISOs rely on frequency measurements that come from GPS-synced network monitoring hardware. Again, high-end industrial stuff, reliable and well-supported, but single-constellation. Ditto WISP PTMP hardware, mostly deployed in the early 2010s. Heck, even the current stuff is still only single-constellation because that's all it needs to be. Thinking about it further, I think your statement applies to cellphones and very little else.
- brohee 7y ago
  My statement also applies to modern PTP servers, e.g. https://www.heoldesign.com/T108-GPS-GLONASS-BEIDOU-GALILEO-Time-server?lang=en https://www.heoldesign.com/T108-GPS-GLONASS-BEIDOU-GALILEO-T... If you are in the market for one, IMHO nowadays multi-constellation should be a baseline requirement.

by brohee 7y ago

zaarn 7y ago

What disaster would knock out only one constellation out of four (perfectly one) and also go unnoticed to users?

brohee 7y ago
Politically driven shutdown. But the point I just made makes it unlikely, it's not the weapon it once was.
jerf 7y ago
The phrase that should strike fear into the heart of any engineer who has to work at scale (any kind of scale, not just software): "Correlated failures". A software bug that affects only one of them. The political possibility mentioned elsewhere. Some computer virus never "intended" for a satellite manages to get into one of them for some reason, most likely a very bizarre one that nobody would have ever guessed could happen but still somehow did. Some hardware component that turns out to fail all at the same time (more or less), used by all of a particular constellation. Or most likely of all, something I can't even think of to put here as an example, but still it somehow turns out that a particular failure ends up having as a side effect greatly increasing the chance of failures in the rest of the system despite all the efforts to prevent that. (The "ones we can't even think of" end up with elevated probabilities precisely because we don't think to prevent them and thus put no engineering into preventing them.) All of these outcomes are low probability, sure, but they are all vastly higher probability than sufficient simultaneous independent failures to take down the system. It's why despite all the incredible software engineering that goes into Google and AWS, they still go down sometimes. Almost no conceivable set of independent events can take those things down, but correlated failures still can, do, and will continue to do so.
Piskvorrr 7y ago
On the contrary, I have a hard time imagining an event that would knock all of them out and was not a relatively minor side effect of a far bigger problem, e.g. global thermonuclear war or a hypermassive solar eruption.

michaelt 7y ago

In theory - but if there's one thing leap seconds and GPS rollovers and suchlike have taught me, it's that once-every-few-years situations aren't always robustly supported even if they're entirely foreseeable.

myself248 7y ago

Maybe most equipment purchased in 2019, from top-tier vendors, with all the software options unlocked. But that's several steps away from what's actually deployed in the world right now. A tremendous amount of the telecommunications network uses BITS sync equipment deployed in the late 90s, because it was extraordinarily well built, extremely reliable, and supported over the long term. There's nothing wrong with it, but it's single-constellation. Ditto the power generation and distribution industry, mostly deployed in the early 2000s. ISOs rely on frequency measurements that come from GPS-synced network monitoring hardware. Again, high-end industrial stuff, reliable and well-supported, but single-constellation. Ditto WISP PTMP hardware, mostly deployed in the early 2010s. Heck, even the current stuff is still only single-constellation because that's all it needs to be. Thinking about it further, I think your statement applies to cellphones and very little else.

brohee 7y ago
My statement also applies to modern PTP servers, e.g. https://www.heoldesign.com/T108-GPS-GLONASS-BEIDOU-GALILEO-Time-server?lang=en https://www.heoldesign.com/T108-GPS-GLONASS-BEIDOU-GALILEO-T... If you are in the market for one, IMHO nowadays multi-constellation should be a baseline requirement.