Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ECC matters

realworldtech.com

Post reply on HN
Print view View original

471–480 of 567 posts

Re: ECC matters

#471

Post by 8note »

Glanford wrote: post #243
mark-r wrote: post #196

Earlier quoted context omitted.

I have a simple way of remembering when to leave out the apostrophe. His, hers, its are all possessive and none of them have an apostrophe.

In this particular case 'it's' can also be possessive although it's considered non-standard, so to be correct you can always treat it like a contraction of 'it is'.

That would be contracting "its is"?

The ambiguity makes it hard for me to say. "It's the best" vs "its is the best"

It doesn't feel right dropping the is

Re: ECC matters

#472

Post by vel0city »

magila wrote: post #385
marcosdumay wrote: post #72

Earlier quoted context omitted.

Keep in mind that Intel lies about its TDP.

There's been a lot of misinformation spread about what TDP means for modern CPUs. In Intel's case TDP is the steady state power consumption of the CPU in its default configuration while executing a long running workload. Long meaning more than a minute or two. The CPU implements this by keeping an exponentially weighted moving average (EWMA) of the CPU's power consumption. The CPU will modulate its frequency to keep…

Intel and AMD absolutely have "turbo mode" and market such features using the term "turbo". It might just be a result of the weighted moving average, but the term "turbo" isn't something reviewers just made up out of nowhere.

https://www.intel.com/content/www/us/en/architecture-and-tec...

https://www.amd.com/en/technologies/turbo-core

Re: ECC matters

#473

Post by hda2 »

incrudible wrote: post #358
dkersten wrote: post #235

Earlier quoted context omitted.

Yes. And? That doesn't suddenly make bitflips benign.

The point is that you can't prevent failure by just buying something. You have to deal with the fact that failure can not be prevented . In other words, if a single defective DIMM somewhere in your deployment is causing catastraphic failure, your mistake was not buying the wrong RAM modules. Your mistake was relying on a single point of failure for mission critical data.

People might not be preventing failure by using ECC, but they significantly decrease the likelihood of having to deal with hard-to-debug problems caused by bitflips.

Re: ECC matters

#474

Post by amluto »

jgalentine007 wrote: post #206
cbanek wrote: post #152

Earlier quoted context omitted.

Well, I've had to actually worry about radiation bitflips as well. It does happen. But usually not so much on Earth!

One of the tire pressure sensors in my car tires had a bit flip a couple years ago and I had to reprogram it's ID. Luckily it was a subaru, so only a light came on in the dash. My old Honda crv however would turn traction control on if your pressure was low - which worked by applying brakes to wheels that were slipping. If you were going up a slippery hill you would soon have no power, sliding backwards nearly off th…

Hah. My oven’s bottom unit stopped working. I strongly suspect a bit flip.

Re: ECC matters

#475

Post by fart32 »

andrewstuart2 wrote: post #414
grishka wrote: post #413

Earlier quoted context omitted.

I remember reading how someone registered some google domains with a single bit flipped, and saw actual requests coming to them.

If you or anybody can remember the source article, that sounds like an interesting read! Edit: found one with a quick search. https://nakedsecurity.sophos.com/2011/08/10/bh-2011-bit-squa... And https://www.researchgate.net/publication/262273269_Bitsquatt...

Also here: https://www.youtube.com/watch?v=gXY3jm34RFU

I can't find it right now, but I also saw someone doing this with some version of Google Now (or what it was back then) and serving different assets into it.

Re: ECC matters

#476

Post by ClumsyPilot »

CydeWeys wrote: post #78
defanor wrote: post #66

Earlier quoted context omitted.

I've considered using an AMD CPU instead of Intel's Xeon on the primary desktop computer, but even low-end Ryzen Threadripper CPUs have TDP of 180W, which is a bit higher than I'd like. And though ECC is not disabled in Ryzen CPUs, AFAIK it's not tested in (or advertised for) those, so one won't be able to return/replace a CPU if it doesn't work with ECC memory, AIUI, making it risky. Though I don't know how common i…

> but even low-end Ryzen Threadripper CPUs have TDP of 180W, which is a bit higher than I'd like. Why does it matter? It doesn't idle that high; it only goes that high of you're using it flat out, in which case the extra power usage is justified because it's giving that much more performance over a 100 W TDP CPU. Now I totally get it if you don't want to go Threadripper just for ECC because it's more expensive , but…

If the concern is about heat and noise, the cooling system is the most important factor, oddly I don't see it even mentioned.

Get a huge cooler like Noctua d14, and you pc becomes silent. It lasts forever, requires no maintenance, a good investment.

If you are adventurous, watercooling is even better, but its a can of worms I decided I'd rather live without - possibility of leaks and cost make it harder to justify

Re: ECC matters

#477

Post by finiteloop »

nostrademons wrote: post #237

I still remember Craig Silverstein being asked what his biggest mistake at Google was and him answering "Not pushing for ECC memory." Google's initial strategy (c. 2000) around this was to save a few bucks on hardware, get non-ECC memory, and then compensate for it in software. It turns out this is a terrible idea, because if you can't count on memory being robust against cosmic rays, you also can't count on the soft…

One of the best quotes in the Google quotes file an early Googler maintained (I am sure I am screwing it up):

“I’ve heard of defensive programming, but never adversarial memory.” — Ben Gomes

Re: ECC matters

#478

Post by xmodem »

lighttower wrote: post #376
cbanek wrote: post #77

As someone who has had to read thousands of random game crash reports from all over the interwebs (you know when Windows says you might want to send that crash log? like that), I totally agree. Of all the things to be worried about, like OS bugs, bad hardware configuration, etc. bad memory is one of those really troubling things. You look at the code and say "it's can't make it here, because this was set" but when yo…

Someone reads those reports!?! Wow, how do I write them to ensure someone who reads them takes them seriously?

The best way is to submit the same crash report from thousands of different locations, repeatedly

Re: ECC matters

#479

Post by itisit »

nostrademons wrote: post #237

I still remember Craig Silverstein being asked what his biggest mistake at Google was and him answering "Not pushing for ECC memory." Google's initial strategy (c. 2000) around this was to save a few bucks on hardware, get non-ECC memory, and then compensate for it in software. It turns out this is a terrible idea, because if you can't count on memory being robust against cosmic rays, you also can't count on the soft…

New Yorker article that credits Jeff Dean and Sanjay Ghemawat with discovering the company’s bitflip issue:

https://www.newyorker.com/magazine/2018/12/10/the-friendship...

Re: ECC matters

#480

Post by fulafel »

KMag wrote: post #453
axiolite wrote: post #448

Earlier quoted context omitted.

> At some point, whatever's watching the watchers is going to be vulnerable to bitflip One advantage of microkernels is that the "watcher" is so small that it could be run directly from ROM, instead of loaded into RAM. QNX has advocated that route for robotics and such in the past. Minix may not be the best example of the type. While it is a microkernel, it's real world reliability has been poor in the past. More mat…

Anyone have any idea why there haven't been any open-source QNX clones, at least not any widely known ones? Even before their Photon MicroGUI patents expired, the clones could have used X11. I used to occasionally boot into QNX on my desktop in college. It was a very responsive and stable system. Hypervisors are, to a first approximation, microkernels with a hardware-like interface. All of this kernel bypass work bei…

There are well known open source microkernels, like Minix 3 and L4. Probably not that attractive.

Why something hasn't been done is always a hard question to answer, since to succeed a lot of things have to go right, and by default none of them do. But one thing is that microkernels were more trendy in the 90s - r&d people are mostly doing things like "the cluster is the computer", unikernel, exokernel, rump kernel, embedded (eg tock), remote attestation since then (I'm not up to date on the latest).

Post reply on HN