Cosmic rays causing 30,000 network malfunctions in Japan each year:
Cosmic rays are causing an estimated 30,000 to 40,000 malfunctions in domestic network communication devices in Japan every year, a Japanese telecom giant found recently.
Most so-called "soft errors," or temporary malfunctions, in the network hardware of Nippon Telegraph and Telephone Corp. are automatically corrected via safety devices, but experts said in some cases they may have led to disruptions.
It is the first time the actual scale of soft errors in domestic information infrastructures has become evident.
Soft errors occur when the data in an electronic device is corrupted after neutrons, produced when cosmic rays hit oxygen and nitrogen in the earth's atmosphere, collide with the semiconductors within the equipment.
[...] Although NTT did not reveal if network communication disruptions have actually occurred, the company said it was "implementing measures against major issues" and "confirming the quality of the safety devices and equipment design through experiments and presumptions."
(Score: 0, Funny) by Anonymous Coward on Thursday April 08 2021, @09:16AM (3 children)
Like Japan doesn't have its own neutrons from their nuke plants, need to blame the aliens for it. Fucking racists
(Score: 0) by Anonymous Coward on Thursday April 08 2021, @01:15PM (2 children)
Neutrons are neutral... they don't interact with anything except other neutrons.
(Score: 0) by Anonymous Coward on Thursday April 08 2021, @03:19PM (1 child)
Neutrons are the majority mass particle in the universe. It's time we recognize that neutron privilege is to blame for everything that ails us physically.
(Score: 0) by Anonymous Coward on Thursday April 08 2021, @04:37PM
Neutralize the neutrons!
(Score: 2) by MostCynical on Thursday April 08 2021, @09:28AM
It doesn't specify which hardware was apparently corrupted - Space hardened routers, anyone?
No extra cost, of course..
Actually, maybe plastic [universetoday.com] will do..
"I guess once you start doubting, there's no end to it." -Batou, Ghost in the Shell: Stand Alone Complex
(Score: 3, Insightful) by Snotnose on Thursday April 08 2021, @09:48AM (1 child)
How many millions of packets get sent/routed/forwarded every hour.
30k errors per year is nothing.
Of course I'm against DEI. Donald, Eric, and Ivanka.
(Score: 1, Funny) by Anonymous Coward on Thursday April 08 2021, @02:30PM
I agr
(Score: 4, Interesting) by VLM on Thursday April 08 2021, @12:19PM (2 children)
Aka probably doing nothing special.
Its pretty trivial as a physics experiment to graph and curve fit cosmic rays vs altitude, although it actually varies by air pressure which varies by altitude. So that's the micro and physics side of the analysis.
I don't know about telcos in Japan but its trivial in the USA to pull the semi-encoded lat lon data from text format layout records. Then mush that against an elevation lookup (northings and westings don't include upwardings or whatever). Then you're already pulling bit error quality data from low level ckts or at the fiber level since at least the 90s because I was there LOL. Then you plot the bit error rate vs altitude on an aggregate scale. Then you can do statistical woo woo to extract the effect of cosmic rays vs rando electrical noise from lightning or intermittent circuits or fiber multipath or whatever other cause.
I've worked in mostly flat areas; there's no point in trying. I imagine a telco guy working in a mountainous area could have fun analyzing this type of data.
The interesting data is the ratio between the cosmic rays and other bit error sources. That's probably not public.
The point of doing all this at a telco is you set thresholds for long term performance monitoring and I would guess the thresholds have to vary by altitude if you work for a mountainous telco otherwise you'll send field techs to prevent maint circuits at the top of mountains all the time and actual intermittent ckts at sea level might get ignored so you'd get more preventative maint bang for buck (or yen...) if you correct your prioritization data for cosmic rays.
(Score: 3, Informative) by Muad'Dave on Thursday April 08 2021, @01:21PM
These guys [spaceweather.com] track cosmic ray radiation from balloon launches and airplane trips regularly. Scroll _way_ down the page. There's also neutron data along the left column.
(Score: 1, Informative) by Anonymous Coward on Thursday April 08 2021, @10:45PM
There have been studies done on "single bit" errors caused by (presumably) cosmic rays. In fact, this is a big part of why we use Error Correction Code (ECC) Memory https://en.wikipedia.org/wiki/ECC_memory [wikipedia.org] to mitigate these effects. The Wikipedia article contains information about the studies
Unfortunately, ECC memory is not generally supported for desktop systems, but it is supported for servers. A pity, because the increase in cost is minimal for the benefit. Intel chipsets do not generally support ECC on the desktop side, but do on the server side. AMD supports it on the server side, and on newer desktops
(Score: 1) by shrewdsheep on Thursday April 08 2021, @03:13PM (1 child)
TLDR; I'm just wondering how they managed to attribute malfunction to cosmic rays. How can they even distinguish software bugs from random bit flips, let alone flips due to cosmic rays? If checksumming of some static RAM regions is used, my criticism would be that it would be first, very specific to certain failures and second, could not prove even a random cause, as software bugs/malicious manipulation cannot be ruled out.
(Score: 0) by Anonymous Coward on Thursday April 08 2021, @03:37PM
They tried to blame it on global warming but their bosses would have none of that.