Hacker News Reader

Comment by gbrown_ | Hacker News Reader

Can't say I really like the title as it comes across as an absolute statement whereas the bit flip could have happened for any number of unknown reasons.

I saw a similar take on Twitter and whilst root causing such things (especially when it's a single occurrence) isn't always possible, shrugging and saying "cosmic rays" should be the last thing to posit not one of the first.

View on HN

"cosmic rays" are more of a well-known term of art for "single bit flipped with unknown hardware cause" than a reference to literal cosmic rays

It is not a "well-known term", it is a cliché.

Clichés are all well known terms. Not all well known terms are clichés.

Interesting, because I was thinking of ... cosmic rays.

sometimes it really is cosmic rays! but you usually can't know after the fact

I remember an older less-computer savvy gentlemen asking for support because "the program isn't working", when after a few questions we realized his computer won't boot up (screen dark, etc.). We thought his terminology was all screwed up. But now I realize he just lacked the necessary PR skill. He should have said that "the program isn't working" is a well-known term of art for power users such as himself. The fact that people who actually have a clue about computing find this imprecise term upsetting if there problem. It so happens that aside from developing software, I'm physicist working on particle physics (specifically, I make my living in industry from cosmic rays). So I can assure you that "cosmic rays" actually mean something. Something very specific. Books have been written full of specific intelligent things we can say about cosmic rays. If you go an appropriate it to describe a whole bunch of phenomena because you can't be bothered to distinguish between them, you're in the exact same boat as that gentlemen from the start of the story. Using wrong terms prevents understanding, as can be seen in all the stories linked elsethread so far. For example, using thinner silicon typically reduces the rate of Single-Event Upsets (the malfunctions caused by cosmic rays) but using smaller silicon components typically increases the rate of malfunctions due to quantum fluctuations. The latter typically happen in specific hardware that we manufactured not-quite-as-well as the rest. SEUs happen in the same rate in all hardware.

You're saying you have to engineer chip components so they're small enough not to be hit as often by cosmic rays, but large enough to avoid quantum fluctuations? Are quantum fluctuations something Intel is dealing with regularly as they get down under 3nm?

Yes, quantum tunneling (an electron "teleporting" into or out of transistors) has been an issue everyone has had to design around for multiple process nodes now.

Go easy on the old guy. It sounds to me like he was completely correct, at the appropriate level of abstraction for him.

And as for cosmic rays...this may be a sensitive topic for you (or maybe you're just in a condescending mood), so I'll tread carefully, but seems simple to me.

The actual cause of the problem, given the instrumentation in place at the time of incident, is unknowable. But it might have been a subatomic particle. It absolutely definitively is, sometimes.

Metonymy might be imprecise, but it's human. I'm not sure what standard you intend to hold commenters here (or old guys with computers) to.

> If you go an appropriate it to describe a whole bunch of phenomena because you can't be bothered to distinguish between them

Cosmic ray is the appropriate colloquial term. Just like "bug" is the appropriate term to describe computer problems that have nothing to do with insects. It's a well established colloquialism and not simply terminology made up on the spot like in your example.

Why? When you do Raman spectroscopy, in any given session you're likely to see 1 going through a 1 cm^2 sample. Students are taught that if you get a super sharp increasing looking peak, first rerun it because you probably just saw a cosmic ray.

Hardware issues that cause bitflips like that one famous issue[1] with Sun servers in ~2000 (supposedly caused by radioactive isotopes in IBM manufactures SRAM chips) are often called "cosmic rays".

1 - https://www.computerworld.com/article/2585216/mcnealy-blames...

> are often called "cosmic rays".

No! This exact issue is a classic example because at one point the fault was attributed to cosmic rays by Sun before the true cause came to light. As another commenter has said, terminology matters.

Reply on news.ycombinator.com