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

Everything I Know About SSDs

kcall.co.uk

Post reply on HN
Print view View original

161–170 of 190 posts

Re: Everything I Know About SSDs

#161

Post by outworlder »

rodgerd wrote: post #119
notacoward wrote: post #2

The part about filesystems is slightly incorrect. > The way the file system handles this is incompatible with the workings of NAND flash. That's true of most conventional filesystems, but log-structured filesystems are much more flash-friendly. That's why there has been a resurgence of interest in them, and also why a typical flash translation layer bears a striking resemblance to a log-structured FS. There are also…

He's also inaccurate in characterizing all cells as "1 or 0"; digital electronics have always been a cut-off of an underlying analogue value (such as voltage).

They explain this in more detail later on. See "SSD Reads" and "interpreting the results".

Re: Everything I Know About SSDs

#162

Post by CrazyStat »

rckoepke wrote: post #50
notacoward wrote: post #2

The part about filesystems is slightly incorrect. > The way the file system handles this is incompatible with the workings of NAND flash. That's true of most conventional filesystems, but log-structured filesystems are much more flash-friendly. That's why there has been a resurgence of interest in them, and also why a typical flash translation layer bears a striking resemblance to a log-structured FS. There are also…

> to an HDD all sectors are the same. Also I believe some humans, and (filesystems?) intentionally stored certain data towards the inside/outside of the HDD because the simple cylinder geometry allowed faster reads in those regions. However, I'm not seeing conclusive proof that modern HDDs show performance variation with respect to radius.

I remember reading a blog post ca. 2005, before SSDs were common, in which the author explained how he had identified all the files his system needed to boot and the order in which they were read from disk, and then contrived to put them all in the correct order in a contiguous block at the start of the drive to eliminate seeks during boot entirely. This apparently shaved a couple seconds off his boot time.

Re: Everything I Know About SSDs

#163

Post by ncmncm »

CamperBob2 wrote: post #146
ncmncm wrote: post #101

Earlier quoted context omitted.

It's pointedly edgy, because thousands of people died. Or 30, if you believe the govt. report.

Either way, far more people die prematurely every day from causes due to pollution from coal-fired plants. But if you're a journalist or documentary filmmaker trying to make something look all cool and edgy and scary and shit, an exploding nuclear power plant makes for a much more interesting story than just another day in the pulmonary ward.

If not for some very edgy heroics, it would have been millions, and most of Ukraine down to the Black Sea uninhabitable for centuries. What did happen was not just "because nuke", but because of a whole series of design and management failures, very far from the least of which was denialism.

They did finally fix the cause of the explosion in the other reactors of that design, remarkably many years later.

Nobody here will defend coal, but graphite-moderated reactors are not the tech you want to be defending.

Re: Everything I Know About SSDs

#164

Post by toast0 »

labawi wrote: post #120
toast0 wrote: post #63

Earlier quoted context omitted.

> How exactly can you control the nuclear decay? That's what the control rods are for. The uranium in one fuel rod in isolation decays at whatever natural rate, which would warm water but not boil it, and placing the rods near each other allows for the decay products (high energy particles) to interact with other fuel rods and induce more rapid decay. The control rods slot in between the fuel rods, and absorb the dec…

> The control rods slot in between the fuel rods, and absorb the decay products without inducing further nuclear decay. Usually these are graphite rods. AFAIU graphite rods increase fission by slowing (not capturing) neutrons which in turn have a better chance of propagating further fission, because .. physics. Quite nifty actually - without the moderator, the fuel wont burn.

I think you're right; I misinterpreted the term 'graphite-moderated reactor' to mean something it doesn't. Graphite will slow the neutrons so they react more. Also, the Chernobyl reactor design has graphite tips on its control rods, which I misremembered as the primary substance of the rod.

The primary substance of the control rods is (usually) a neutron absorber, and most reactors with control rods have a passive safety system, so gravity and springs will force the control rods in to significantly slow the reaction unless actively opposed by the control system.

The Chernobyl rods had graphite ends so that when fully retracted, the reactor output was higher than if there was simply no neutron absorber present; unfortunately, this also meant that going from fully retracted to fully inserted would increase the reactivity in the bottom of the reactor before it reduced it, and in the disaster, this process overheated the bottom of the reactor, damaging the structure and the control rods got stuck, and then really bad things happened.

Long story short, most control rods don't have graphite. ;)

Re: Everything I Know About SSDs

#165

Post by gergoerdi »

TremendousJudge wrote: post #71
winrid wrote: post #31

Earlier quoted context omitted.

Not OP but in my experience people tend to have trouble comprehending systems of planetary gear sets. Although modern automatics are probably a bit easier to understand than old ones, especially CVTs? As long as you're ok with "the computer just triggers this solenoid.." rather than understanding a big hydrualic computer.

Modern automatic transmissions are actually manual transmissions with a robot moving through the gears as far as I know. However, they do a bunch of stuff that I don't understand like pre-engage the next gear so the switch is faster -- I have no idea how that works

I've found that building the Lego Technic Porsche 911 set from a couple years ago really helps a lot with understanding dual-clutch transmission systems. https://brickset.com/sets/42056-1/Porsche-911-GT3-RS

Re: Everything I Know About SSDs

#166

Post by red_phone »

Marsymars wrote: post #65
blattimwind wrote: post #4

Earlier quoted context omitted.

In practice all SSDs are always encrypted because they use the encryption to whiten the data written to them. That's why "Secure Erase" takes less than a second on SSDs, it doesn't erase anything but the key.

Yes, but in practice, this encryption isn't competently implemented. Hence why MS stopped supporting hardware-accelerated BitLocker. https://twitter.com/SwiftOnSecurity/status/11774296582599270...

They still support it, it’s just not the default any longer for drives that report the capability.

Re: Everything I Know About SSDs

#168

Post by SlowRobotAhead »

wtallis wrote: post #160
SlowRobotAhead wrote: post #152

Earlier quoted context omitted.

So stay under 8 physical NVME and it should be fine?

A typical NVMe SSD has a four-lane PCIe link, or 2+2 for some enterprise drives operating in dual-port mode. So it usually only takes 2 or 3 drives to saturate an 8-lane bottleneck. Putting 8 NVMe SSDs behind a PCIe x8 controller would be a severe bottleneck for sequential transfers and usually also for random reads.

I need to think about this for a second.

You’re saying the performance gains stop at two drives in raid striping. RAID10 in two strip two mirror would still bottleneck at 8 total lanes?

I also need to see about the PERC being limited to 8 lanes - no offense - but do you have a source for that?

Edit: never mind on source, I think you are exactly right [0] Host bus type 8-lane, PCI Express 3.1 compliant

https://i.dell.com/sites/doccontent/shared-content/data-shee...

To be fair; they have 8GB NV RAM, so it’s not exactly super clear cut how obvious a bottleneck would be.

Re: Everything I Know About SSDs

#169

Post by kalleboo »

mehrdadn wrote: post #151
wtallis wrote: post #150

Earlier quoted context omitted.

It stands to reason. The low LBAs get filled first and used most. The outside of the disk is where the linear speed is fastest, and it's closest to where the heads park.

Yeah I understood that immediately after I read it, but it's not how typical optical discs work, and I'd just assumed they work the same way until now. Hadn't really thought about it before. What I find a bit bizarre though that some defragmenters have an option to move files to the end of the drive. Wouldn't you want to move files to the beginning in that case?

Maybe if you're optimizing for write performance of new files (e.g. back in the day when sequential disk performance was more of a bottleneck, you'd want lots of contiguous free fast disk space for something like digitizing video)

Re: Everything I Know About SSDs

#170

Post by Marsymars »

red_phone wrote: post #166
Marsymars wrote: post #65

Earlier quoted context omitted.

Yes, but in practice, this encryption isn't competently implemented. Hence why MS stopped supporting hardware-accelerated BitLocker. https://twitter.com/SwiftOnSecurity/status/11774296582599270...

They still support it, it’s just not the default any longer for drives that report the capability.

Thanks for the correction - I expect that support is just a temporary state of affairs though. I used hardware accelerated BitLocker for a few years, and had my systems break multiple times due to BitLocker-related regressions when MS pushed updates. I can't imagine it's going to get more attention now that's it's never enabled by default.
Post reply on HN