Live data from Hacker News

Cosmic rays causing 30k network malfunctions in Japan each year

mainichi.jp

81–90 of 120 posts

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#81
post #24

Earlier quoted context omitted.

I'm speculating here but it would be more likely that software bugs cause more errors than cosmic rays.

Found the rust programmer.

The rust programmer is the one out there ending that phrase with "and this a problem that must be solved".

...

Well, the rust programmer and the nuclear engineer, possibly.

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#82

Kind of a fun ancedote: In "the old days" when you'd run a server room in your office we had a very large HPC cluster + a significant amount of storage and other one off servers on the top floor of a mid rise office building. I eventually moved it all to a former nuclear fallout facility where our systems were three floors down and under a gigantic pool of water. Error rates and random crashes fell off IMMEDIATELY. I…

Can you give OOM of any of the improvements? Eg. crash rates halved?

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#83

Earlier quoted context omitted.

Have a source? That seems like an awesome way to recycle grocery bags.

Water is also a good protector. And the polyethylene is actually a proxy for "hydrogen atoms". The reason is that dense hydrogen has a lot of protons to interact with the incoming radiation. But unfortunately plastic bags are not dense polyethylene. You would kind of need full blocks of solid polyethylene...

You can make solid polyethylene out of plastic bags by pressing them in a mold heated to ~100-150C.

But it's quite flammable, so you might not want to use it as a building material.

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#84
post #33

Earlier quoted context omitted.

The peer comment posted a source that has lots of references. Though most of them are tests in space, where I assume the thickness requirements would rule out grocery bags. I am curious how thick a layer of HDPE you would need on earth to make any notable difference.

That linked nature paper seems to indicate 10g/cm^2 for a 50% reduction. A standard shopping bag is about 5g, so you would need roughly 20k bags/m^2, or ~2k bags/ft^2. At 0.9g/cm^3, that would be roughly 11cm of solid polyethylene

Ah, so bags are out, as is practically the ceiling. But cut sheet HDPE is easy to find, so tiles atop your machine would be relatively cheap and easy.

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#85
post #22

Kind of a fun ancedote: In "the old days" when you'd run a server room in your office we had a very large HPC cluster + a significant amount of storage and other one off servers on the top floor of a mid rise office building. I eventually moved it all to a former nuclear fallout facility where our systems were three floors down and under a gigantic pool of water. Error rates and random crashes fell off IMMEDIATELY. I…

Sun’s UltraSPARC II CPU modules were claimed to be particularly susceptible to cosmic rays circa 2000.

Not so much susceptible as "came with their own radioactive source"

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#86
post #79

Earlier quoted context omitted.

Wonderful. Yet another reason for the IT office to be relegated to a sub-basement.

Why the whole IT office? Just the hardware running the software, not the humans who develop it.

I feel like OP was joking

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#87
post #39

Earlier quoted context omitted.

What is the tolerance for cosmic rays for a human... I ask because on the journey to Mars you’ll get quite a few I guess and also on the planet’s surface after you’re there with such a thin atmosphere?

This is actually a thing people are criticizing Musk for, he is being accused of sending people to Mars with insufficient shielding against cosmic rays. In my opinion it's a bit silly argument, as there's a whole bunch of other risks and quality of life sacrifices made by the persons who are going to undertake that journey. Some raised chance of cancer is probably the least of their worries.

He can't send people to Mars yet so how are people accusing him of sending people to mars with insufficient shielding? There are limits to criticism of Musk one would think.

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#89

Kind of a fun ancedote: In "the old days" when you'd run a server room in your office we had a very large HPC cluster + a significant amount of storage and other one off servers on the top floor of a mid rise office building. I eventually moved it all to a former nuclear fallout facility where our systems were three floors down and under a gigantic pool of water. Error rates and random crashes fell off IMMEDIATELY. I…

And people still think the sun is nuclear. Ha!

Re: Cosmic rays causing 30k network malfunctions in Japan each year

#90

Ahh, the ol' Cisco get-out-of-jail-free card. Headline would be more accurate if it said "30K unidentified bugs blamed on unverifiable phenomenon each year in Japan".

Incorrect. According to the article NTT experimented with neutrons (simulated cosmic rays) in a lab.

NTT systematically set out to be able to distinguish these different kinds of error, and you go ahead and ignore that effort.

Post reply on HN