Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

SSD Latency, Error Rates May Spell Bleak Future

timothy posted more than 2 years ago | from the everything-counts-in-large-amounts dept.

Bug 292

Lucas123 writes "A new study by the University of California and Microsoft shows that NAND flash memory experiences significant performance degradation as die sizes shrink in size. Over the next dozen years latency will double as the circuitry size shrinks from 25 nanometers today, to 6.5nm, the research showed. Speaking at the Usenix Conference on File and Storage Technologies in San Jose this week, Laura Grupp, a graduate student at the University of California, said tests of 45 different types of NAND flash chips from six vendors using 72nm to 25nm lithography techniques showed performance degraded across the board and error rates increased as die sizes shrunk. Triple-Level NAND performed the worst, followed by Multi-Level Cell NAND and Single-Level Cell. The researchers said MLC NAND-based SSDs won't be able to go beyond 4TB and TLC-based SSDs won't be able to scale past 16TB because of the performance degradation, so it appears the end of the road for SSDs will be 2024."

cancel ×

292 comments

Sorry! There are no comments related to the filter you selected.

Sounds legit (5, Insightful)

sbrown7792 (2027476) | more than 2 years ago | (#39065855)

Because there could *never* be a breakthrough discovery/invention found within the next 10 years.

Re:Sounds legit (5, Funny)

jedidiah (1196) | more than 2 years ago | (#39065885)

OK then. You've got 10 years. Get going.

Re:Sounds legit (5, Funny)

fuzzyfuzzyfungus (1223518) | more than 2 years ago | (#39066401)

Tiny monks with tiny paintbrushes, inscribing ones and zeros on individual electrons. No problem.

Re:Sounds legit (5, Funny)

mickwd (196449) | more than 2 years ago | (#39066759)

Would those be chipmonks?

OK, OK, I'm going........

Re:Sounds legit (4, Insightful)

mcavic (2007672) | more than 2 years ago | (#39067049)

Well, to start with you can make an SSD as big as you want by taking smaller SSD's and chaining them together with an intelligent front-end.

Re:Sounds legit (5, Insightful)

Anonymous Coward | more than 2 years ago | (#39065953)

Oh there will be a great discovery/invention in the next 10 years. Unfortunately it will be tied up in patent litigation for the next 50 years after that. All fun and games when it is a hard drive. Not so funny when it is a medicine that can save your kid.

Re:Sounds legit (1)

RandomAvatar (2487198) | more than 2 years ago | (#39066273)

That is only if it isn't raised during that 50 years...

Re:Sounds legit (5, Funny)

LearnToSpell (694184) | more than 2 years ago | (#39066727)

Won't somebody think of the hard drives!

Re:Sounds legit (2, Insightful)

sonicmerlin (1505111) | more than 2 years ago | (#39066871)

HDD tech has advanced without patent litigation tying anything up. What makes you think it will be different for NAND's successor?

Re:Sounds legit (4, Interesting)

Grishnakh (216268) | more than 2 years ago | (#39066079)

We already have the breakthrough, but it's not Flash, it's PRAM [wikipedia.org] .

Re:Sounds legit (5, Funny)

bughunter (10093) | more than 2 years ago | (#39066153)

Yes, but what I heard about PRAM is that you have to push it. A lot.

Re:Sounds legit (5, Funny)

Anonymous Coward | more than 2 years ago | (#39066233)

But I *LIKE* to push the PRAM a lot!

Re:Sounds legit (0)

tscheez (71929) | more than 2 years ago | (#39066235)

awesome ^^^^

Re:Sounds legit (0)

Anonymous Coward | more than 2 years ago | (#39066927)

ok, I give, someone explain it.

Re:Sounds legit (1)

Anonymous Coward | more than 2 years ago | (#39067073)

It's a reference to Monty Python and the Holy Grail.
Turn in your geek card.

Re:Sounds legit (5, Informative)

maxwell demon (590494) | more than 2 years ago | (#39066255)

We already have the breakthrough, but it's not Flash, it's PRAM [wikipedia.org] .

And MRAM. [wikipedia.org] And FeRAM. [wikipedia.org]

Re:Sounds legit (1)

Anonymous Coward | more than 2 years ago | (#39066083)

You mean like this [eetimes.com] ?

Re:Sounds legit (4, Interesting)

Zouden (232738) | more than 2 years ago | (#39066093)

Perhaps it's already been found:
http://en.wikipedia.org/wiki/Phase-change_memory [wikipedia.org]
PCM still has hurdles to overcome, but it's generally considered that performance increases as size decreases, the opposite of NAND.

Re:Sounds legit (1)

Anonymous Coward | more than 2 years ago | (#39066597)

For mobiles, currently phase-change is a little slow, not available in large quantities and is more like RAM than storage which impacts the software stack a bit. It's already in some low-end mobiles (you'd be surprised just how low, but it's an excellent NOR replacement) but it's not suitable for smart phones.

At the same time, current eMMC (which is fast becoming the only option on newer chipsets) is getting too slow and too unreliable to hit a predicted 5-year use pattern.

Re:Sounds legit (1)

westlake (615356) | more than 2 years ago | (#39066115)

Because there could *never* be a breakthrough discovery/invention found within the next 10 years.

Tomorrow's "breakthrough" doesn't mean you have a commercially viable product within the next ten years --- or even the next twenty,

Re:Sounds legit (4, Insightful)

peragrin (659227) | more than 2 years ago | (#39066271)

what about the last 3-4 years worth of discoveries, of phase change, memresistors, etc. Many of which get more efficient the smaller you go.

So NAND Flash has a lifespan. big deal, So did magnetic core drives, Hard drives are still going strong but are reaching the top ends for themselves too.

Re:Sounds legit (4, Insightful)

jimbolauski (882977) | more than 2 years ago | (#39066473)

NAND has been around for 14 years and they are trying to extrapolate out to 2024, almost double it's life span. I'm trying to think of any technology that was 10 years old that there was a road map of where it would be in another 10 that turned out to be accurate.

Re:Sounds legit (2)

sgt scrub (869860) | more than 2 years ago | (#39066133)

Wasn't there about 6 alternatives to NAND discovered last year? I think IBM announced 2 of them.

Re:Sounds legit (1)

0racle (667029) | more than 2 years ago | (#39066263)

So obviously research into the limits of the current technology is pointless?

Re:Sounds legit (1)

Bengie (1121981) | more than 2 years ago | (#39067015)

Memresistors are suppose to come out in the next 1-2 years. Will be even used as system memory because it has no effective "wear". That "breakthrough" is already done, it's just being readied for production.

Re:Sounds legit (3, Interesting)

PopeRatzo (965947) | more than 2 years ago | (#39067021)

Because there could *never* be a breakthrough discovery/invention found within the next 10 years.

Didn't you hear? We've reached the limitations of technology and innovation.

That's why it's so stupid to put any money into non-fossil energy. If we can't power a house by solar energy now, we'll never be able to and we just have to accept it.

It's the End of History. Again.

Re:Sounds legit (0)

Anonymous Coward | more than 2 years ago | (#39067119)

just friggin mega-raid 'em? seems to me that 4tb+ solutions using regular hd's smaller than 1tb do scale. and of course you will be able to integrate the tech that makes them purr into smaller units during that time too(which is pretty much what the specialised chips talked about in the article, which the researchers didn't use, are already doing - they're filling that exact need already today).

of course another thing is that processing silicon would get cheaper too, so keeping the chips the same area(and just scaling the process) wouldn't be an absolutely necessary either(though as the processes get more refined, the error rates get down as well as the technology to make that happen matures and so the quality of the chips will go up - the most immature technology had the highest error rates in the research, doh..).

they sure had nice toys for the research though. doesn't really make that research too useful.

HDDs for the win! (0, Troll)

Anonymous Coward | more than 2 years ago | (#39065857)

SSD = fail

Can we finally throw this garbage to the heap now?

Re:HDDs for the win! (4, Funny)

Surt (22457) | more than 2 years ago | (#39065999)

Yes, please send your SSDs to me for disposal, thanks.

SSD =/= NAND Flash (5, Informative)

MischaNix (2163648) | more than 2 years ago | (#39065879)

There will be other solid-state storage solutions. The only reason NAND is currently used is its relative cheapness and reliability.

Save us, MRAM! (1)

Anonymous Coward | more than 2 years ago | (#39065883)

You heard me.

In other news... (4, Informative)

Troyusrex (2446430) | more than 2 years ago | (#39065887)

An old study (well, executive) showed that there was a world wide demand for "maybe 6" computers. This might all be true at current technology levels but technology will have changed an awful lot by 2024.

Re:In other news... (1)

jedidiah (1196) | more than 2 years ago | (#39065929)

....conflating what's possible with what's desired.

One of these depends on human nature and the other one depends on physics.

nature always finds a way (1)

Thud457 (234763) | more than 2 years ago | (#39066873)

Moore's law has been extended to the point where it is a sociological or economic observation, not a physical one.

Re:In other news... (1)

Desler (1608317) | more than 2 years ago | (#39065959)

Yeah a quote that has no actual historical evidence to back up its legitimacy. Want to try again?

Re:In other news... (5, Insightful)

dyingtolive (1393037) | more than 2 years ago | (#39066025)

"Hey, would you want a computer? It's a city block large, uses all of these punchcards for I/O, and doesn't really do much other than crack Enigma. Hey, where are you going?"

"Hey, would you want a computer? It can fit in your pocket, let you talk to anyone in the world, can take pictures and provide you god damn near any information written down by a human being, and you can watch porn on it!"

Computers are the same thing they were even 20 years ago in name only.

Re:In other news... (-1)

na1led (1030470) | more than 2 years ago | (#39066107)

Bill Gates said - "640K ought to be enough for anybody".

Re:In other news... (3, Funny)

Anonymous Coward | more than 2 years ago | (#39066279)

A year? Yeah, I could live pretty well off of that...

Re:In other news... (2)

Lussarn (105276) | more than 2 years ago | (#39066283)

Actually, there is no credible source on that...

Re:In other news... (1)

kryliss (72493) | more than 2 years ago | (#39066711)

Hopefully I'll have my hoverboard by then.

Stuff like this... (4, Insightful)

blahplusplus (757119) | more than 2 years ago | (#39065935)

... always denies other areas of innovation. The same way processors were thought not to scale down to x nm and we're at 20'ish nm now. The same way hard drives were thought only to have x capacity and we're now in the terabytes. If nand is really so limited then something different then nand will take it's place. But a few terabyte will be more then enough for 99% of applications and hard disks will be for packrats and those who need large amounts of longer term storage.

Re:Stuff like this... (1)

Anonymous Coward | more than 2 years ago | (#39065995)

640k ought to be enough for everyone!

Re:Stuff like this... (5, Insightful)

blahplusplus (757119) | more than 2 years ago | (#39066145)

"640k ought to be enough for everyone!"

One can take a look at videocards, right now for most PC gamers they haven't needed to upgrade their video hardware for quite some time relatively speaking compared to the past. The idea that needs will scale linearly forever is nonsense.

There is a point after rapid growth where you reach 'good enough' until the next step is ready which no one knows in advance.

Re:Stuff like this... (0)

Anonymous Coward | more than 2 years ago | (#39066289)

His points still stands. Saying it will be enough is rather short-sighted. In the future, we might have full blows androids (no not the phones etc), with real AI walking about. I can't even guess at how much processing power, RAM, and HD space one of those would need.

Re:Stuff like this... (1)

jedidiah (1196) | more than 2 years ago | (#39066677)

That's funny because I just tried to install a Linux RTS game on a couple of my other machines. Neither one of them was equipped to deal with it. "Better drivers" on Windows didn't help the Win7 box. I didn't even try the Mac; it was trailing edge when it was new.

Good enough changes year by year and people find new ways to exploit hardware that some people might have thought is "tapped out".

VDPAU and VAAPI are both good examples of that.

I would still ditch pretty much any embedded GPU on the market today and replace it with something less lame.

Re:Stuff like this... (0)

Anonymous Coward | more than 2 years ago | (#39067003)

"I would still ditch pretty much any embedded GPU on the market today and replace it with something less lame."

And guess what I said in my prior post? I said VIDEOCARDS (as in add in high performance video cards). I don't think you have been following videocards the last few years at all, the performance gaps between video cards on the PC have shrunk. Not to mention the dwindling performance of every new generation of cpu's due to multi-core bottlenecking dependent code of which there is significant amounts in most applications. The new 7970 that just came out is barely faster then a GTX 580 in real games. The performance of cpu's and hardware has been slowing down and not only that the software you run on this hardware and tools to create it are significantly lagging behind because most games are developed for consoles/multiplatform hardware target and/or mobile/ipad.

If we're going to be honest lets talk about the software, since hardware is nothing without a justification for it's purchase. The last 10 years of gaming have slowed to a crawl in terms of innovation, in fact most modern games are so expensive and time consuming to make the quality of most games has stagnated because of it when you take away the cinematics that they use to compensate for lack of interesting gameplay. Buying new hardware for the last little while has become more and more difficult to justify given the barren landscape of killer apps. This bottleneck isn't away anytime soon.

Re:Stuff like this... (0)

Anonymous Coward | more than 2 years ago | (#39066179)

64nm ought to be enough for everyone! ;-)

Re:Stuff like this... (1)

Anonymous Coward | more than 2 years ago | (#39066291)

You just keep telling yourself that to get through the lonely nights.

Re:Stuff like this... (0)

Anonymous Coward | more than 2 years ago | (#39066901)

Damage control by the vendors of the rotating media....... Gah... Sticks in the mud.. Who wants to wait for rotational and positioning latency for a block of data??? Bah.. Death to rotating media.

Re:Stuff like this... (1)

PRMan (959735) | more than 2 years ago | (#39067045)

One time, my wife's cousin (who was studying RAM at MIT and is now a brain specialist teaching at Stanford) said that "you will never be able to put more than 40MB on a PCMCIA card."

I replied, "Within 5 years, we'll be carrying a GB around in our pocket the size of a postage stamp." I was right. Sometimes smart guys are so focused on their area that they fail to see the realities of supply and demand combined with Moore's law.

no one will ever (-1)

Anonymous Coward | more than 2 years ago | (#39065943)

need more than 4TB

billy g.

Yep, and it's impacting mobile device design *now* (1)

Anonymous Coward | more than 2 years ago | (#39065949)

This has been a pain heading down the pipeline for cellphone designers for a couple of years now, and its not like you can still buy stuff built on the old processes either unless you ordered it a couple of years ago and I only know one cellphone company confident enough to pre-order a few years of storage.

Storage is getting larger, whoopee. It's also becoming massively slower and less reliable. Two attributes you can't really make up for in volume on a phone. Even so-called 'high end' SLC eMMC is miles slower than the raw NAND we used to be able to get. You get 10x as much now, but you can only use it 1/20 as long.

LCD's also started like that (1)

Anonymous Coward | more than 2 years ago | (#39065951)

Blah blah, with years of revenue, disks will become cheaper because the production lines, investment risks and other burdens of innovation implementation will be paid for. Same thing happened with LCD production, and we get better LCDs now than we used to have.

Re:LCD's also started like that (1)

Desler (1608317) | more than 2 years ago | (#39066015)

Except that older flash is actually much better than the more recent multi level kind. So it's the opposite of LCD.

Obligatory ... (1, Redundant)

techstar25 (556988) | more than 2 years ago | (#39065985)

"16TB ought to be enough for anybody."

Re:Obligatory ... (1)

AliasMarlowe (1042386) | more than 2 years ago | (#39066683)

"16TB ought to be enough for anybody."

Didn't you mean "640TB blah blah blah".
We've already got at least 20TB of fixed disks at home (including online backups). The media server alone has 12TB.

I want HAL's memory (3, Interesting)

na1led (1030470) | more than 2 years ago | (#39066001)

Still waiting for the Holographic Memory that should have been hear a decade ago.

Re:I want HAL's memory (2, Funny)

Anonymous Coward | more than 2 years ago | (#39066037)

Holographic memory requires fusion power.

Re:I want HAL's memory (1)

maxwell demon (590494) | more than 2 years ago | (#39066305)

Holographic memory requires fusion power.

Not true. A warp core powers it just fine.

Re:I want HAL's memory (4, Funny)

MobyDisk (75490) | more than 2 years ago | (#39066563)

A warp core really isn't a power source. It is more like an alternator. The power source is the matter-antimatter reactions. Similarly people confuse dilithium crystals with being a power source when they are really just a matter-antimatter regulator.

And now, back to reality...

Re:I want HAL's memory (2)

freeze128 (544774) | more than 2 years ago | (#39067109)

I really think of dilithium as more of a catalyst than a regulator. It turns the explosion of a matter/antimatter reaction into highly energetic plasma that can be readily used in many applications. Then, the acoustic phase compensators even out the flow of plasma so it's more fluid and not as.... What? Star Trek has been off the air for years? Aw, crap! Nevermind....

Re:I want HAL's memory (5, Funny)

Anonymous Coward | more than 2 years ago | (#39067113)

So... How's your virginity going?

Re:I want HAL's memory (2)

SmurfButcher Bob (313810) | more than 2 years ago | (#39066963)

Not a problem, since I have that in my goddamned flying car.

Re:I want HAL's memory (3, Funny)

roman_mir (125474) | more than 2 years ago | (#39066617)

Still waiting for the Holographic Memory that should have been hear a decade ago.

- there is the problem.

With holographic memory you shouldn't be trying to 'hear' anything, it's something likely in visible electromagnetic spectrum instead!

will vertical 3D NAND flash be here in 2013 (1)

cyfer2000 (548592) | more than 2 years ago | (#39066011)

as reported here [semi.org] and here [nature.com] ? I thought people have been busy about it for quite some time.

"...the end of the road for SSDs will be 2024..." (5, Insightful)

John Hasler (414242) | more than 2 years ago | (#39066087)

Yes. They'll all stop working then and it will become impossible to make any more.

Re:"...the end of the road for SSDs will be 2024.. (2)

Surt (22457) | more than 2 years ago | (#39066363)

Well, not so much that but rather than hard drive rotational latencies will finally catch up to nand. With our disks spinning at a paltry 100,000,000 rpm, latency will finally be a worry of the past.

Re:"...the end of the road for SSDs will be 2024.. (1)

rtaylor (70602) | more than 2 years ago | (#39066713)

I've wondered what spinning disk could do in a vacuum chamber and with a non-contact magnetic bearing.

Not bleak at all (2, Informative)

Anonymous Coward | more than 2 years ago | (#39066123)

From the article, "This will reduce the write latency advantage that SSDs offer relative to disk from 8.3x (vs. a 7 ms disk access) to just 3.2x.". Yeah, doom and gloom.

Re:Not bleak at all (4, Informative)

Joehonkie (665142) | more than 2 years ago | (#39066217)

Exactly, this quote at the end says it all: "However, even with TLC flash at 6.5nm, Grupp calculates that SSDs will continue to outperform hard disk drives on throughput, 32,000 IOPS to 200 IOPS, respectively."

Throughput isn't measured in IOPS (1)

addikt10 (461932) | more than 2 years ago | (#39067041)

Actually, that's when I realized that the guy writing the article didn't have a clue. Since when is throughput measured in IOPS?

I'm sure it's all wonderful (5, Funny)

goldcd (587052) | more than 2 years ago | (#39066167)

But I'm choosing to ignore it all, entirely based on font.
http://cseweb.ucsd.edu/~lgrupp/CV.pdf [ucsd.edu]

Man we are toast (in 12 years..) (1)

bobbied (2522392) | more than 2 years ago | (#39066185)

Surely somebody will come up with a new way to store data in the next 12 years. Just a hunch, but I'm thinking that somebody is going to dust off some old research or idea and we are all going to be rushing out to replace our flash drives for something faster long before we start hitting this limit. This sounds like some researcher just threw down the gauntlet. Innovation will prove that his assumptions where incorrect because their conclusion looks like ones I've heard in the past.

Anybody recall the "640kB out to be enough for anybody" quote?

One thing does stand out though, 2 TB limits on NAND flash drives? Really? 16TB on TLC? Given standard drive sizes seem to be multiplying by 2 about every year, there will be serious issues even before the 2024 date.

Re:Man we are toast (in 12 years..) (2)

Surt (22457) | more than 2 years ago | (#39066427)

No, no one remembers it. It's an invention of fantasy, not memory.
http://en.wikiquote.org/wiki/Bill_Gates#Misattributed [wikiquote.org]

Re:Man we are toast (in 12 years..) (2)

jedidiah (1196) | more than 2 years ago | (#39066761)

The funny thing is that this "myth" was well established long before he published any sort of rebuttal.

This is ancient history. It happened decades ago and finding evidence now would be difficult even if you knew where to look. Chances are that any such corroboration faded away by the time that rebuttal was published.

4TB limit (4, Insightful)

afidel (530433) | more than 2 years ago | (#39066205)

Yeah, about that 4TB limit, I think these [fusionio.com] folks will be surprised that their 5TB and 10TB drives won't be possible in the next few years....

Re:4TB limit (2)

Desler (1608317) | more than 2 years ago | (#39066355)

I hate to break it to you but that is 8 drives in one device. Hence the "octal" name.

Re:4TB limit (0)

Anonymous Coward | more than 2 years ago | (#39066485)

Any device that has more than one flash die will satisfy your definition of multiple drives in a device.

This is all besides the point: the article is talking about single flash chips.

Re:4TB limit (1)

Desler (1608317) | more than 2 years ago | (#39066789)

That was my point. He was trying to disapprove the quote by using a device that does not disprove the quote since it doesn't use 1 chip.

Re:4TB limit (0)

Anonymous Coward | more than 2 years ago | (#39066891)

Well, those iodrives aren't "8 drives in one" any more than my intel sata SSD is 6 drives in one. Octal is just a name, they have dozens of flash chips.

Re:4TB limit (1)

Surt (22457) | more than 2 years ago | (#39066455)

They're talking about on a single device. Those drives are arrays of something like 64 devices.

Re:4TB limit (2)

blueg3 (192743) | more than 2 years ago | (#39066561)

It's the size of a single double-wide PCI card. Okay, scratch that, it *is* a single double-wide PCIE card. That counts as a single device. Just like how if you put a bunch of hard drive platters behind a common interface within a standard-size hard drive shell, it counts as one hard drive.

Re:4TB limit (1)

amicusNYCL (1538833) | more than 2 years ago | (#39066501)

They probably cheated by putting more than 96 NAND dies in their device. 96 NAND dies should be enough for anyone. (sorry)

Re:4TB limit (1)

rtaylor (70602) | more than 2 years ago | (#39066627)

That's not exactly a single chip.

Re:4TB limit (1)

Anonymous Coward | more than 2 years ago | (#39066881)

But it points out the FUD of the article summary.

so it appears the end of the road for SSDs will be 2024

This is, of course, complete bullshit.

CPUs are not obsolete despite having hit significant limits of single chips; we simply have moved to multi-chip and multi-core solutions.

Video cards are not obsolete; you can freakin' hook up veritable arrays of them on the high end.

Conventional hard drives are not obsolete, though I know of no one, for storing anything remotely important, who relies on a single drive.

End of the road? We're already moving to multi-device-in-a-device with solid state. The end of the road is not apparent and cannot be predicted by a flippant Slashdot summary.

Re:4TB limit (1)

bloodhawk (813939) | more than 2 years ago | (#39067079)

It doesn't point out the FUD of the article at all, that is like saying 720k floppy disk storage is not obsolete as I can create a drive that takes 10TB of them. but yes the summary is garbage. It is also important to remember that NAND is not the only SSD technology and hence the death of NAND based SSD does not equate to the death of SSD.

Re:4TB limit (0)

afidel (530433) | more than 2 years ago | (#39066999)

The researchers used PCIe-based flash cards with a channel speed of 400MBps based on the Open NAND Flash Interface (ONFI) specification and a standard 96 NAND flash dies, which is typical of SSDs.
...
By the time NAND flash shrinks from 25nm today to 6.5nm in 2024, SSDs based on TLC flash will sport as much as 16TB of capacity and MLC flash SSDs will have 4TB, Grupp said.


So basically they use a generic PCIe design and then concluded that PCIe designs wouldn't exceed 4TB by 2024 whereas I showed that a PCIe device, available in the market today, has greater capacity than they claim will be the ceiling in 12 years. It is utter and complete horseshit.

Re:4TB limit (1)

phizi0n (1237812) | more than 2 years ago | (#39066837)

Those are pci-e cards, not standard form factor 5.25" drives. The OP simplified the researcher's conclusions which were actually more along the lines of "as die size decreases, so does performance." Manufacturers can throw more chips on a drive to improve performance without shrinking die size but if they try to use smaller die sizes then performance of the chip, both latency and throughput, decreases. If they want to keep drives a certain phsyical size while also increasing capacity then they need to shrink the die size but that reduces performance - eventually to the point that a standard HDD will perform better.

Just add more (2)

phyr (586855) | more than 2 years ago | (#39066257)

Can't scale past 16TB? Why not just stack them?

Re:Just add more (4, Informative)

Surt (22457) | more than 2 years ago | (#39066503)

It costs money to stack. At a much higher rate than it does to scale. Or at least that has been the case. It will be a significant hit to the industry when they can no longer count on device scaling to help bring up density, and get forced to wire multiple chips in ever expanding arrays.

Did Anobit have anything to do with this? (1)

alen (225700) | more than 2 years ago | (#39066259)

I only ask because Apple is the largest flash customer/reseller in the world and they just bought this company

Really, who thinks this stuff up. (0)

Anonymous Coward | more than 2 years ago | (#39066393)

The future seems to be predicted by a moron with a telescope peering into the sun... I do mean that as well.

So, if die size says that getting smaller is more error prone, current die size and manufacturing techniques can continue just fine. The are benefits to shrinking the die but when those fail you don't just simply lose the old methods. Current SSD pricing isn't about cost to manufacture but rather innovation cost. So at some point it is possible to stack a metric crap ton of current existing chips into very large drives, assuming they don't want to just expand on the current die size.

This article assume there are no big breakthroughs in the coming years. I know it's been a while since we seen a huge one (while being 2 years).

increasing the size of SSD drives is about the same as increasing a raid 0 array. Just add more.

Re:Really, who thinks this stuff up. (0)

Anonymous Coward | more than 2 years ago | (#39066995)

increasing the size of SSD drives is about the same as increasing a raid 0 array. Just add more.

Yes, and, like increasing a RAID 0 array, decreases the MTBF -- when one of the problems highlighted is increasing failure rates.

You need RAID 10 -- RAID 5 would work for integrity, (since SSDs fail in different ways and for different reasons than spinny rust, and the recovery process is more likely to be "read everything and dump to a brand-new SSD" than "replace one chip and rebuild the old SSD", the odds of a second failure during recovery is quite low), but SSDs are for speed, so you're not going to want the parity-writing overhead.

Think of the children! (0)

Anonymous Coward | more than 2 years ago | (#39066593)

How can anyone, in good conscience, bring a child into this world knowing they will only be able to get 4TB SSDs?

Re:Think of the children! (1)

Skapare (16644) | more than 2 years ago | (#39066763)

Because, maybe, one of those children will innovate, and solve this problem?

Re:Think of the children! (1)

jedidiah (1196) | more than 2 years ago | (#39066781)

I will be happy (but not satisfied) when just a mere 1TB SSD does not cost the same as a house or a (decent) car.

Flash retention times (1)

Guppy (12314) | more than 2 years ago | (#39066699)

Along with error rates, what will happen to retention times as the cell size shrinks?

Supposedly, flash memories have expected retention times as short as 5-10 years or so (if not refreshed by re-writing), thanks to gradual leakage of the trapped charges they use to record data; this value is expected to drop as flash cells get smaller. I've had gadgets whose firmware mysteriously become corrupted after sitting around for a few years, and sometimes they could be revived by re-flashing them -- I sometimes wonder if this kind of retention problem could have been responsible.

Do they still use unbalanced conductors in ... (1)

Skapare (16644) | more than 2 years ago | (#39066797)

... microelectronics fabrication, making them vulnerable to inductive effects?

NAND Successor (1)

sonicmerlin (1505111) | more than 2 years ago | (#39066847)

I was kind of hoping we'd have something better than NAND Flash within 5 or so years. Maybe something using memristors? NAND is just too expensive to be useful. Prices haven't dropped in a couple years.

I'm ok withh this (0)

Anonymous Coward | more than 2 years ago | (#39066973)

By 2024 are we still going to be computing the same way? 12 years ago we were terrified of the Y2K bug and smatphones were a pipe dream.

The performance advantages of SSDs and declining costs make them well worth the investment. They aren't going anywhere. With the expansion of tablet computing as an entertainment and leisure medium its still worth it to invest in SSD technology.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>