×

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!

Is SSD Density About To Hit a Wall?

Soulskill posted more than 3 years ago | from the right-into-a-brick-wall dept.

Data Storage 208

Zombie Puggle writes "Enterprise Storage Forum has an article contending that solid state disks will stay stuck at 20-25nm unless the materials and techniques used to design Flash drives changes, and soon. 'Anything smaller and the data protection and data corruption issues become so great that either the performance is abysmal, the data retention period doesn't meet JEDEC standards, or the cost increases. Though engineers are working on performance and density improvements via new technologies (they're also trying to drive costs down), these are fairly new techniques and are not likely to make it into devices for a while."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

208 comments

The cure is the memristor (4, Informative)

symbolset (646467) | more than 3 years ago | (#33622898)

Memristor [bbc.co.uk] technology doesn't even work with feature sizes that big, so it's the logical next step. Also it can be layered and so leverage Dimension Z. Products expected in three years from a joint HP and Hynix venture. No worries.

Or more likely PCM (3, Informative)

Wesley Felter (138342) | more than 3 years ago | (#33623096)

HP and Hynix are doing memristors, while the entire rest of the industry is doing phase-change memory.

Re:Or more likely PCM (0)

Anonymous Coward | more than 3 years ago | (#33623114)

HP and Hynix are doing memristors, while the entire rest of the industry is doing phase-change memory.

Ford is making motorized horseless carriages while the entire rest of the industry is making horse and buggy carriages.

Re:Or more likely PCM (1)

cheesybagel (670288) | more than 3 years ago | (#33623236)

Hynix is the second larger DRAM manufacturer after Samsung or something. I suspect they are a major player in NAND flash as well.

Re:Or more likely PCM (1)

ghjm (8918) | more than 3 years ago | (#33624150)

Phase-change memory offers as much or more promise as memristors for flash-type storage. Yes, memristors are also interesting for thousands of other applications and PCM isn't, but that doesn't make it a horse and buggy for this application.

Re:Or more likely PCM (0)

Anonymous Coward | more than 3 years ago | (#33623172)

And let's not forget MRAM [everspin.com]. Yes, it is indeed quite far from current Flash densities.

Re:Or more likely PCM (5, Funny)

owlstead (636356) | more than 3 years ago | (#33623200)

Yeah, they like to push the P-RAM a lot.

Re:Or more likely PCM (0)

Anonymous Coward | more than 3 years ago | (#33623520)

I have no mod points, and I must approve.

Re:Or more likely PCM (4, Interesting)

cheesybagel (670288) | more than 3 years ago | (#33623230)

Phase-change memory... Oh dear. I still remember when it was being pushed as Ovonic Unified Memory (OUM) or calcogenics. I certainly hope Samsung and the usual suspects can get this to work. But it has been a long time in coming. Well, maybe not as long as MRAM but still...

Re:Or more likely PCM (3, Interesting)

TheRaven64 (641858) | more than 3 years ago | (#33623540)

PC-RAM stands a good chance of being the long-term future (I had the good fortune recently to share a very nice bottle of port with one of the scientists behind underlying technology, and came away quite convinced, and a lot drunk), but the largest currently shipping PC-RAM modules are 64MB. It has a lot of catching up to do before it reaches, let alone passes, the density of flash.

Re:Or more likely PCM (2, Funny)

Jah-Wren Ryel (80510) | more than 3 years ago | (#33623588)

came away quite convinced, and a lot drunk

I think you'll find that's not quite so good an argument in favor of the technology once you've sobered up.

Re:Or more likely PCM (1)

commodore64_love (1445365) | more than 3 years ago | (#33623936)

>>>PC-RAM stands a good chance of being the long-term future

I don't see Flash Drives replacing disk drives, anymore than I see Cartridges making a comeback in gaming.

Re:The cure is the memristor (1)

Tumbleweed (3706) | more than 3 years ago | (#33623688)

Also it can be layered and so leverage Dimension Z.

Beware of Dimension Z...I heard Red Lectroids can come through there. Not worth the risk.

Re:The cure is the memristor (0)

Anonymous Coward | more than 3 years ago | (#33624506)

Memristors are a cross between fusion (always 20-50 years away, but in reality no actual progress has ever been made) and holographic storage -- doable in some form, but too expensive in reality, even though we have had companies from Tamarak and newer deliver drives which were unfeasibly expensive.

I wish people would stop bothering to bring memristers up. Just because some guy at HP discovered one material with memristor-like properties in an experiment doesn't mean that this means it can be turned into something useful, much less marketable dense storage in any meaningful period of time. Just like having one NP border in a lab somewhere doesn't mean one can suddenly wire up a 710 million transistor i7 chip the next day and have it ready for mass fabbing the week after.

Maybe in 20-30 years, but right now it is is still in the "gee, we saw something, lets make a theory about it" stage. Compare this to SSDs which are an actual working technology that is seeing actual duty cycles.

Re:The cure is the memristor (0)

Anonymous Coward | more than 3 years ago | (#33624820)

It's either this, or the only other viable alternative: Biologic Ultra Transistor memory, BUT-RAM.

So... (4, Insightful)

dcmoebius (1527443) | more than 3 years ago | (#33622920)

Improving upon current SSDs will require new technology! Isn't that sort of implied in the whole concept of, you know, progress?

Re:So... (2, Interesting)

mikehoskins (177074) | more than 3 years ago | (#33623102)

Agreed. And, I believe that 34nm is near the best they can do today, in any kind of production.

So, if you can go from a 34nm * 34nm feature to a 20nm * 20nm feature, you can almost triple the density.

So, in the same space you can produce a 128G drive, you can then produce a roughly 384G drive, going from 34nm to 34nm.

So, if a USB Keychain is produced w/ 128G, a 384G can be produced at the same size, barring other issues.

That assumes they are even using 34nm process SSD's, today, to produce 128G USB SSD drives. If they are using a 40nm process, then expect 512G USB SSD's, as a future possibility.

This doesn't even take into consideration stacking SSD vertically and horizontally in a RAID configuration on a drive and maximizing use of space (packaging, support chips, etc.) or making larger physical USB devices.

In the future, hardware compression, deduplication, etc., may further add to storage improvements.

My best guess? 1 Terabyte uncompressed on a keychain, eventually, assuming a 20nm process.

If they can go further than 20nm or improve in other ways, all the better.

Re:So... (2, Informative)

vadim_t (324782) | more than 3 years ago | (#33623142)

512GB SSDs aren't a "future possibility"

1TB SSDs already exist [ocztechnology.com]

Re:So... (2, Informative)

liquiddark (719647) | more than 3 years ago | (#33623168)

That's a bit of a trick, though. They're effectively putting several drives in a specialized RAID package.

Re:So... (4, Informative)

AllynM (600515) | more than 3 years ago | (#33624508)

*EVERY* SSD is a 'specialized RAID package'.

Allyn Malventano, CTNC, USN
Storage Editor, PC Perspective

Re:So... (1)

OffTheWallSoccer (1699154) | more than 3 years ago | (#33623886)

> Agreed. And, I believe that 34nm is near the best they can do today, in any kind of production.

There are at least two companies manufacturing 25nm parts right now.

The wall, and the end of the world. (3, Insightful)

durrr (1316311) | more than 3 years ago | (#33622926)

The wall or plateu or whatever you prefer to call it of electronics progress is similar to the recurring doomsday predictions. It's always right around the corner, but it never happens.
I guess we could liken it to fusion, strong AI, the second coming of Jesus and whatever else that generally is put in the belive it when see it folder.

Re:The wall, and the end of the world. (3, Insightful)

MichaelSmith (789609) | more than 3 years ago | (#33622968)

The wall or plateu or whatever you prefer to call it of electronics progress is similar to the recurring doomsday predictions. It's always right around the corner, but it never happens.

It has to happen.

Re:The wall, and the end of the world. (3, Funny)

durrr (1316311) | more than 3 years ago | (#33623150)

Well obviously the earth will be fried when the sun goes red giant. However I'm quite certain though that the year 5 billion AD equivalent of a electronics engineers will sit in jovian orbit, hellbent on the continuation of moores law and waiting for the sun to turn white dwarf so they can get to work with their new fancy sub-nm electron-degenerate matter lithography techniques.

They'll wake you up from cryo when they're done just to taunt you: "Oh, we're having a few billion years between the nodes now, but it fits the curve, just as i told you."

Re:The wall, and the end of the world. (1)

RicktheBrick (588466) | more than 3 years ago | (#33623902)

They were mass producing 1 Gbit ram chips in 1999. It is now almost 2011 or almost 12 years later. It seems to me if Moore's had continued we should be talking about the 1 Tbit ram chip by now. I think there is a definite wall with ram memory

Re:The wall, and the end of the world. (1)

mlts (1038732) | more than 3 years ago | (#33624778)

Very true. However, what happened is that computing dealt with that plateau by finding ways around it. Caching comes to mind for this. After caching in RAM (main RAM and DRAM on the controllers) comes tiered storage and using faster drives as cache for swap (think ReadyBoost.)

If one tier of computing (l1 cache, RAM, storage) doesn't expand, another will. RAM densities have not gone up that much, but hard disk densities have, so a lot of work is put into caching. If by chance we end up with a breakthrough in RAM densities that make it possible to have 100Tbit DRAM chips, there will be a mechanism that writes data to non-volatile storage coupled with batteries to ensure the DRAM arrays stay refreshed until the writes are completed. If by another chance, we end up with a breakthrough allowing CPUs to have millions of registers on a die without any real architecture impact, we will see CPU microcode to take advantage of that.

Re:The wall, and the end of the world. (1)

MichaelSmith (789609) | more than 3 years ago | (#33623980)

They'll wake you up from cryo when they're done just to taunt you: "Oh, we're having a few billion years between the nodes now, but it fits the curve, just as i told you."

I can't wait. If anybody can keep Moore's law going for five billion years then good luck to them.

Re:The wall, and the end of the world. (1)

ScrewMaster (602015) | more than 3 years ago | (#33624950)

They'll wake you up from cryo when they're done just to taunt you: "Oh, we're having a few billion years between the nodes now, but it fits the curve, just as i told you."

I can't wait. If anybody can keep Moore's law going for five billion years then good luck to them.

I'm not sure I'd want to meet a computer that was the result of five billion year's worth of Moore's Law. Of course, if the Universe is cyclic then so far as we know, a race that existed in some previous incarnation of the Big U might have built a machine that kept evolving itself long after its creators were dust. Hell, it might even have become God, or something so close that we'd never be able to tell the difference.

Re:The wall, and the end of the world. (1)

mlts (1038732) | more than 3 years ago | (#33624598)

It does happen and we find a way around it. Take CPUs. We ran into a wall with clock speed, so we are going with more cores. Once adding tons of cores onto dies stops giving tangible returns, we might go with stacking larger and larger caches, or some future 3D masking technology to allow the caches to be stacked on top of or below the rest of the CPU. When that peters out, there is always moving to 128 bit word lengths, adding more registers, and even newer CPU architectures and emulation.

Re:The wall, and the end of the world. (1)

rm999 (775449) | more than 3 years ago | (#33622996)

There are notable counterexamples. For example, CPU clock speeds have been approaching a limit for years now. The only reason computers get "faster" over time is Moore's Law, which allows the CPU to do more per clock.
http://www.gotw.ca/publications/concurrency-ddj.htm [www.gotw.ca]

Re:The wall, and the end of the world. (0)

Anonymous Coward | more than 3 years ago | (#33623024)

der uh what ?

Re:The wall, and the end of the world. (1)

IB4Student (1885914) | more than 3 years ago | (#33623074)

Limit on clock speeds for silicon is 40 Ghz.

Re:The wall, and the end of the world. (3, Informative)

vadim_t (324782) | more than 3 years ago | (#33623120)

That's not practical though.

At that speed, the signal will travel about 0.6cm per clock cycle. Even at current clock rates at least one clock cycle will pass while the signal simply travels to the RAM chip on the motherboard, without accounting for any circuitry, just the time spent on the wire.

Re:The wall, and the end of the world. (1, Interesting)

Anonymous Coward | more than 3 years ago | (#33623220)

The current cpu architecture will be ditched and will go asynchronous.
L1 L2 and L3 cache will become a conscious thing for high performance applications
The only thing we will see the end of are the half hearted abstractions that make no sense, programmers should know at the cpu level what their code does.
However, hardware abstractions are fine, and would allow many many threads to use every bit of that 40Ghz speed
CPUs could go back to the 6500 simplicity and we would have thousands of them at 40Ghz and 64KB of on die cache per processor
Then maybe a few terabytes at the L2 level and a few thousand TB of main memory
The operating system would just be a hypervisor to manage the loading of all the processes on all of the processors
There is no wall, there is only a wall in ones perceptions. A very simple processor @ 20Ghz with thousands of cores would kick ass

Re:The wall, and the end of the world. (1)

monkyyy (1901940) | more than 3 years ago | (#33623222)

and when that becomes a issue they will start hard wiring ram in to the cpu

Re:The wall, and the end of the world. (1)

Khyber (864651) | more than 3 years ago | (#33623592)

What do you think an on-die cache is? Dedicated RAM for the core to process instructions.

Re:The wall, and the end of the world. (1)

Pulzar (81031) | more than 3 years ago | (#33623872)

At that speed, the signal will travel about 0.6cm per clock cycle. Even at current clock rates at least one clock cycle will pass while the signal simply travels to the RAM chip on the motherboard, without accounting for any circuitry, just the time spent on the wire.

So? Nobody says the signal has to get there in one clock cycle. You already have interfaces like PCIe gen 3 where the signal travel time from one pin to another is far larger than one clock period.

Re:The wall, and the end of the world. (0)

cheesybagel (670288) | more than 3 years ago | (#33623246)

There is always diamond or superconductors. Besides the transistors and the wires are not made of silicon anymore. Today there are low-k materials, high-k materials, metal gates, copper wires, etc.

Re:The wall, and the end of the world. (0)

Anonymous Coward | more than 3 years ago | (#33623868)

Uh, dumbass? Those are insulators and wires. Of course the fucking transistors are made of silicon, doofus.

Re:The wall, and the end of the world. (1)

ScrewMaster (602015) | more than 3 years ago | (#33624956)

Uh, dumbass? Those are insulators and wires. Of course the fucking transistors are made of silicon, doofus.

Politeness is a virtue, shithead.

Re:The wall, and the end of the world. (1)

0111 1110 (518466) | more than 3 years ago | (#33623340)

Limit on clock speeds for silicon is 40 Ghz

Maybe for silicon, but for Intel/AMD CPUs the limit seems to be around 3.3 Ghz. Unless you overclock. Then you can expect anywhere from about 4 to 4.5 Ghz with good air or water cooling or 5+ Ghz with phase change. Didn't the Intel roadmaps from around 10 years ago predict 6 or 7 Ghz CPUs by now? Python programmers may want to start brushing up on their assembly skills. Unless there is some kind of major scientific breakthrough single threaded apps and/or non-parallel tasks won't be getting much faster. Luckily GPUs, RAM, and hard drives are still moving ahead as usual. Well, hard drives are slightly stalled at the moment with some manufacturers not even offering their high capacity drives at 7200 RPM (*cough* Samsung *cough*) and others charging a huge premium just to get the same rotational speeds that we have been getting since about 1992.

Re:The wall, and the end of the world. (0)

Anonymous Coward | more than 3 years ago | (#33623498)

Well, hard drives are slightly stalled at the moment with some manufacturers not even offering their high capacity drives at 7200 RPM (*cough* Samsung *cough*) and others charging a huge premium just to get the same rotational speeds that we have been getting since about 1992.

But while you may have been able to achieve rotational speeds of 7200 rpm in 1992, sequential and random transfer speeds have increased many tens of times over. Random access time has not changed much, but that can't be helped.

Re:The wall, and the end of the world. (2, Interesting)

hairyfeet (841228) | more than 3 years ago | (#33623572)

While it is true single threaded apps won't be getting much faster, the big breakthrough is thanks to lowering prices multicores will be the mass market and you can run a lot more single threaded apps at the same time. Hell the 2.8Ghz AMD quad I'm typing this on has 8Gb of RAM and a Tb of HDD space as well as a Gb on the GPU, and the whole thing cost less than my old P4 rig did and it was pretty bottom of the line at the time. Then you look at the roadmap for the AMD Bobcat/Bulldozer line coming next year and we are talking true multithreaded multicores with individual integer lines for each thread, the amount of apps one can run at the same time is gonna be just crazy.

So as an old greybeard that remembers when his VIC20 cost more in today's dollars than an average gamer rig does now, when flash memory would cost you over $100 to hold a couple of floppies worth of data, and when HDDs were measured in Mb, the incredible price drops that are making this incredible power cheap and available to the masses is what is exciting to me. Honestly is anybody gonna care if their 2Tb SSD portable is slightly less than the size of their candybar phone? As long as you can drop it in a pocket or purse I think the bigger thing will be it affordable to the masses like the flash sticks are now. After all they can come up with the most whizz bang tech in the world, but if the only ones that can afford it are a handful of uber-rich and those working admin at the fortune 500 it really isn't gonna change most folks lives, now is it?

Re:The wall, and the end of the world. (3, Interesting)

X0563511 (793323) | more than 3 years ago | (#33623882)

I'd like to see stuff start getting tougher.

When that 2Tb SSD can fall 4 stories (while in use) and carry on without even noticing, then I start getting tingles...

Re:The wall, and the end of the world. (1)

Raenex (947668) | more than 3 years ago | (#33623972)

Didn't the Intel roadmaps from around 10 years ago predict 6 or 7 Ghz CPUs by now?

Try 10 Ghz [geek.com], and even that was scorned at the time for being conservative. It was quite an amazing ride from the earliest CPUs, but alas, it did come to an end. The comments on that page, a blast from the past, are priceless.

Re:The wall, and the end of the world. (1)

nacturation (646836) | more than 3 years ago | (#33623132)

The only reason computers get "faster" over time is Moore's Law, which allows the CPU to do more per clock.

So if Gordon Moore didn't state his law, CPUs would be forbidden from getting faster?

Re:The wall, and the end of the world. (1)

X0563511 (793323) | more than 3 years ago | (#33623890)

So if someone didn't write down the law of Gravity, we would all be an amorphous cloud of particles?

Laws put an observed truth down into words. Whether or not this is done doesn't effect that the observed truth is still there.

Re:The wall, and the end of the world. (1)

Kjella (173770) | more than 3 years ago | (#33623060)

Agreed. If anything I think we will be running into that wall slowly, simply by not having the year-over-year improvements like we're used to. I'd be very surprised if Intel just issues a press release one day that said "You know that 18nm process tech we've been working on? we've hit a brick wall and it's not going to happen."

Re:The wall, and the end of the world. (1)

monkyyy (1901940) | more than 3 years ago | (#33623252)

"You know that 18nm process tech we've been working on? we've hit a brick wall and it's not going to happen." so we have starting looking into shaping cpu`s into cubes with 5 fans around them

Re:The wall, and the end of the world. (1)

mentil (1748130) | more than 3 years ago | (#33624642)

That would only work if the center were a vacuum. Which would be pretty cool, until your CPU implodes.

Re:The wall, and the end of the world. (1)

Arancaytar (966377) | more than 3 years ago | (#33623196)

You mean cold fusion; fusion itself is quite feasible (though not yet able to be harnessed safely).

Re:The wall, and the end of the world. (0)

Anonymous Coward | more than 3 years ago | (#33623288)

DNF is actually expected next year.

Re:The wall, and the end of the world. (1)

Surt (22457) | more than 3 years ago | (#33623758)

In none of those previous cryings of wolf had the number of atoms in a single device dropped into the double digits.

Re:The wall, and the end of the world. (1)

ScrewMaster (602015) | more than 3 years ago | (#33624928)

The wall or plateu or whatever you prefer to call it of electronics progress is similar to the recurring doomsday predictions. It's always right around the corner, but it never happens. I guess we could liken it to fusion, strong AI, the second coming of Jesus and whatever else that generally is put in the belive it when see it folder.

A more logical comparison would be the repeated assertions that hard drives would be reaching their "theoretical maximum" capacities we've all heard for the past couple decades. Now the things are beyond a terabyte and still increasing. Heck, I remember back in the 70's when I was playing around with some 1 kilobit RAM devices that some scientists were predicting that memories wouldn't get much denser than that.

Does it matter? (3, Insightful)

MikeFM (12491) | more than 3 years ago | (#33622936)

It doesn't seem a big deal to me. I'd be more interested in seeing the prices drop and to have larger RAM caches.

Re:Does it matter? (0)

Anonymous Coward | more than 3 years ago | (#33622972)

Dropping prices are due to die shrinkage directly related to transistor size / width (what is I guess what they mean with 20-25nm?).

Ya (1)

Sycraft-fu (314770) | more than 3 years ago | (#33623006)

The only way a density wall will matter is if price can't scale. If price scaling is purely based on density, then we have a problem as SSDs aren't price competitive yet. However if price can keep scaling down, then no problem. As it stands you can pack in flash to easily meet the same density as magnetic storage. They have 512MB SSDs that are 2.5" form factor and I suspect it could be smaller, it is just that size to fit in regular laptop drive slots.

So we'll see. I don't know what are the price barriers for SSDs, and I'm sure smaller lithography helps, but I suspect they'll be fine without it.

Re:Ya (1)

PAjamian (679137) | more than 3 years ago | (#33623162)

They have 512MB SSDs that are 2.5" form factor

Really? I would think it would be more like 512GB ... or more (especially considering the 16GB USB flash memory stick I have)

Re:Does it matter? (1)

vadim_t (324782) | more than 3 years ago | (#33623016)

Probably not much.

In my experience, disk space isn't nearly as limiting as it used to be. Back in '93, a 500MB drive was pretty large, but could be easily filled. I remember that deciding what to keep on one's hard disk, and how to free up a bit more space used to take a considerable amount of time. After all, a single CD was bigger. Today, a 500GB drive won't be filled by most people.

Also, there are 1TB SSDs in existence already, one reported to be postage stamp sized. That's a very useful size, considering that much smaller ones can be comfortably used.

And if it goes get cramped, there are always regular hard disks to fall back to, which do storage of mass amounts of data that doesn't need high performance better anyway.

Re:Does it matter? (2, Insightful)

MikeFM (12491) | more than 3 years ago | (#33623130)

My laptop has a 128GB SSD which would be really cramped except I keep most my files on my NAS where it can be kept in RAID and be automatically backed up etc. Really the local drive should only be the files needed to boot and hook to the network and the rest used to cache the files you're most likely to need soon. As you said you can already get decent storage space in the usual form factors so it's not really a big deal if the drives can't get more dense. I dont really care if my NAS takes up a whole server rack. It's only a matter of time before the cloud handles most our storage anyway. Local storage is just to much hassle for most people anyway and why should we ever worry about a hard limit to how much space we have available?

clouds mean rain (4, Insightful)

Anonymous Coward | more than 3 years ago | (#33623398)

Local storage is a lot cheaper and faster for most people in the USA, which is all I can speak of. Maybe over in Utopialand where everyone has 100 gig speed connections and hosting is pennies a day for terrabytes the "cloud" might be cheaper and better. Our domestic broadband speeds and prices are not even close to keeping up with increased local storage density and lowering prices for same. Saying the "cloud" will do everything is sorta naive, we have all the major ISPs talking about limits and caps now. This is 100% the WRONG time to be shifting to far away "cloud" storage for most people.

  I know I'll be keeping my movies and files handy right here, thanks. I just can't see storing multiple gig sized movies way over there someplace when it would cost me two cents to store it here and have it playback at fast streaming speeds for the cost of the electricity.

Having to go pay yet again to watch your movie or access your own file..nope. The "cloud" is a marketing buzzword for companies that want to charge you serious coin for access to *your own files*.

Density halt, so work on price (3, Insightful)

markdavis (642305) | more than 3 years ago | (#33622994)

Well, the density is already not bad, so the big key is to get the cost down! For larger applications of Flash memory(like over 250GB) I don't think the physical size is going to be a problem because it is competing with 3.5" and 2.5" hard drives.

Aside from cost, there are plenty of other non-density things to work on: number of rewrite cycles, speed, reliability, etc. I can't wait for the day that spinning media eventually goes bye-bye.

Re:Density halt, so work on price (1, Insightful)

Anonymous Coward | more than 3 years ago | (#33623144)

This has to be a joke, right? Semiconductor cost is directly related to the size (amount) of silicon required, and the cost of flash memory is directly related to and by far the most expensive component of an SSD. It has nothing to do with the size of the packaging.

Almost all price decreases for flash memory so far have come from feature size shrinks. The only other large decrease was the implementation of MLC, and that's only happened once so far. X3 and X4 cells have potential but also some pretty severe drawbacks at the moment, other improvements tend to be marginal rather than 50%+ cost decreases.

Re:Density halt, so work on price (1)

markdavis (642305) | more than 3 years ago | (#33623400)

Size/density is only one factor in overall cost. It is an important factor, but there are others, such as economy of scale (diluting overhead through quantity), efficiency (of production), quality/precision (failure-free), and production methodologies (lower overhead through innovation).

As another point, I would argue that size of the packaging is directly related to the size of the silicon.

Re:Density halt, so work on price (0)

Anonymous Coward | more than 3 years ago | (#33624848)

Once built it's necessary to keep a fab continuously close to capacity apart from tooling changeover to even break-even, so no matter what you're still going to need to be producing the same 20,000 wafers per month and obtaining similar economies of scale. The only way to increase volume meaningfully is to build another fab, which is an enormous undertaking and hence happens on a well-defined schedule. Yield and process optimisation should always be sought no matter what and can be anticipated to gradually improve through a process lifecycle - all these things are also very nice but they're a pittance compared to the cost benefit from a process shrink. If you decrease the required fabrication area this will decrease your per-bit cost, and it is by far the fastest route to lowering the price of an SSD.

Flash memory modules are a standard package for the ease of PCB design, availability of BGA pins, pick-and-place tools and equivalence/exchangeability and are in no way related to the actual size of the flash IC. They're mostly empty space and with package-on-package packaging can contain an almost arbitrarily large amount of storage on a single module (I think Toshiba may be leading with 512Gbit/64GB at the moment). The exception would be for highly-constrained form-factors like microSD which is actually almost all IC, for SSDs it's a non-factor.

Re:Density halt, so work on price (1, Insightful)

Anonymous Coward | more than 3 years ago | (#33623156)

*sigh* why is everybody so excited about getting rid of spinning media? It's not that expensive (especially compared with SSDs), and with accelerometers, it pretty much makes up for the issues that are caused by its mechanical nature. I also haven't seen significant power savings from SSDs or significant improvements in computer performance with their use.

I really don't understand the excitement about getting rid of a tried and true technology that still serves today's needs in an inexpensive and reliable manner.

Adobe Flash, on the other hand, needs to go away ASAP.

Re:Density halt, so work on price (1)

markdavis (642305) | more than 3 years ago | (#33623374)

Because compared to solid-state memory, spinning media is not very dense, uses a lot of power, is fairly unreliable, gives off more heat, and is fragile per impact.

Re:Density halt, so work on price (0)

Anonymous Coward | more than 3 years ago | (#33623636)

Odd, I consider none of those as a big issue. I switched to ssd to get rid of the noise and for the speed.

Re:Density halt, so work on price (2, Insightful)

im_thatoneguy (819432) | more than 3 years ago | (#33623856)

Size to a large extent is the cost.

If they can fit twice as much product onto 1" square for the same price then you get an effective decrease in cost.

The incremental costs of memory are somewhat linear. If size was of no concern then they could sell a 10TB drive for the same price if they just put more spindles into one drive. The cost is per spindle. So yes they could sell you a 10TB drive instead of a 1TB drive that was 10x as large physically, but it's going to cost 10x as much.

Good riddance to Flash... (3, Informative)

KonoWatakushi (910213) | more than 3 years ago | (#33623040)

There are far better technologies waiting to replace it, one being P-RAM. The best thing is, none of the newer tech is subject to Flash's crippling block-erase semantics, and so they are far more suitable for SSDs. No longer will SSDs require tremendously complex controllers and firmware in order to attain good performance, allowing new SSDs to be both cheaper, faster, and more reliable.

Re:Good riddance to Flash... (1)

Klinky (636952) | more than 3 years ago | (#33623208)

Right but when is P-RAM going to be available & have the same production and supply chain that NAND has at this point? It's going to take some time going from the 8MB PRAM chips shipping for mobile phone usage to get something similar to the 1TB NAND SSDs we have now. It's similar to OLED displays, where OLED keeps hovering in the background and each year it's poised to replace LCDs, but yet there still aren't any viable consumer level OLEDs on the market.

Who needs more than 8GB in a micro-sd package? (1, Informative)

Anonymous Coward | more than 3 years ago | (#33623044)

Can you imagine a stack of 128 8GB micro-SD cards next to a 1TB 3.5" hard disk? Do we need higher densities? Of course there's a relation of price to die area, but there's still room for improvement in manufacturing technologies. As long as the price keeps going down, I'm fine with the current density. Worry about metrics that matter.

Re:Who needs more than 8GB in a micro-sd package? (0)

Anonymous Coward | more than 3 years ago | (#33623254)

or better yet, 32GB microsdhc cards. Those things are dense. And what size are they using? 34nm?

I realize that there are some major differences between SD cards and SSDs, but they aren't exactly utilizing all of the space in even 2.5" drives. Many of today's SSDs are also offered in 1.8" drives.

"Unless the materials and techniques change" (0)

Anonymous Coward | more than 3 years ago | (#33623062)

"Enterprise Storage Forum has an article contending that solid state disks will stay stuck at 20-25nm unless the materials and techniques used to design Flash drives changes"

Uh huh. And how is this situation different than every single computer technology we've had since 1970?

Buzzzzzap... (1)

3seas (184403) | more than 3 years ago | (#33623088)

..times up.

Guess we will be stuck where we are on this matter forever.

Or does this mean there is something that is going to replace this tech anyway?

Just lower the manufacturing cost (2, Informative)

backslashdot (95548) | more than 3 years ago | (#33623112)

Densities are fine. The main problem is lowering the cost. They need to drop the price by an order of magnitude. I am sure it costs way less than that to manufacture .. they just have to pay back all the research and equipment capital costs and build more production lines. Once they do that it will be dirt cheap. I remember when LCD monitors were a couple thousand bucks. And hard drives were far more expensive than SSDs are today .. and that was only 15 years ago.

For example an OCZ Technology 250 GB SSD is $450 .. I paid around $400 for a 400 Megabyte drive in 1995. That's works out to hard disks back then being nearly 5 times the price per megabyte of SSD drives today.

Re:Just lower the manufacturing cost (1)

backslashdot (95548) | more than 3 years ago | (#33623116)

What I mean by densities are fine is not that we don't need more denisty .. of course we do .. but right now it's price that's the most urgent need.

Re:Just lower the manufacturing cost (2, Interesting)

queazocotal (915608) | more than 3 years ago | (#33623174)

'Just'.
It really does cost quite a lot to make flash.
For example, a fab capable of the latest geometries will set you back over a billion dollars.

This fab is only cutting edge for a yearish before needing retooled, or moving down the value chain to make cheaper - less profitable - stuff.

Re:Just lower the manufacturing cost (3, Insightful)

my $anity 0 (917519) | more than 3 years ago | (#33623386)

But, if the technology hits a brick wall, the fab won't need retooling because there will be no cutting edge technology, increasing the amount of years of useful life, and eventually lowering price. It's slower than the lowering by developing better tech, though, one would assume.

Re:Just lower the manufacturing cost (1, Insightful)

forkazoo (138186) | more than 3 years ago | (#33623248)

Densities are fine. The main problem is lowering the cost. They need to drop the price by an order of magnitude. I am sure it costs way less than that to manufacture .. they just have to pay back all the research and equipment capital costs and build more production lines.

Density = Cost

The more bits per cm^2 of silicon, the less silicon you need to buy in order to store your stuff. When people talk about density, they aren't talking about the physical size of the consumer SSD product, they mean density of feature size on the silicon chips. It's expensive to refine, process, and manufacture super high end silicon. That's why flash tends to scale quite linearly with storage size in a way that mechanical drives never have. The price really is dominated in a large way by the amount of silicon, which dictates the number of bits. To drive down the price, put more bits on less silicon, and you have a winner.

Other options, like figuring out how to use shittier silicon in cheaper fabs to drive down price are also worth some R+D, but density is the proven method for driving down the price.

Si prices (2, Informative)

AlpineR (32307) | more than 3 years ago | (#33623550)

I don't think so. Back when I used to do research on microelectronic fabrication methods, we bought 3-inch wafers for about $10 apiece. Those were high purity with doping to whatever type and level we selected. And that was without bulk pricing or favorable price scaling with larger wafers.

Our molecular beam growth chamber, however, cost hundreds of thousands of dollars plus tens of thousands per year for supplies and maintenance (plus tens of thousands for a postdoc and a grad student to run it).

So I really think the cost of equipment and processing far outweighs the cost of the silicon wafers. Otherwise, all CPU's with the same physical size would have roughly the same price, regardless of transistor count or clock speed.

Re:Just lower the manufacturing cost (1)

fnj (64210) | more than 3 years ago | (#33623600)

Your math is a bit suspect. That's 555 times more, not 5 times more!

Re:Just lower the manufacturing cost (0)

Anonymous Coward | more than 3 years ago | (#33624884)

By my calculations it's over 9000!

Re:Just lower the manufacturing cost (0)

Anonymous Coward | more than 3 years ago | (#33624710)

You don't realize how manufacture's reduce cost do you? It costs about the same to manufacture a 45nm wafer as it does a 34nm wafer as its does a 2Xnm wafer. But as the size goes down, they get more GB per wafer. Thus you can't reduce cost unless you shrink the die. So, its high cost, small density or low cost and high density.

Sure it might hit a wall... (4, Interesting)

gman003 (1693318) | more than 3 years ago | (#33623128)

but who says the wall is going to win that collision? I've seen it time and time again: a problem is encountered, and dealt with. Optical disk rotation speed. Parallel data buses. Processor clock speeds. They all hit a wall, and we got around that wall. We lowered the wavelength of the laser instead of go to 56x CDs. We switched to serial buses when parallel encountered clocking issues. We switched to multicore processors when we couldn't keep upping the gigahertz. I'm fully confident we'll figure out a solution to this problem as well, whether it be new manufacturing techniques, memristors, or just larger Flash chips.

Re:Sure it might hit a wall... (0)

Anonymous Coward | more than 3 years ago | (#33623420)

Larger chips is one pretty interesting idea for low-power electronics where the cooling system is not (yet) the bottleneck of the system.

Another possibility may be to simply stack these low-power chips on top of one another.

Re:Sure it might hit a wall... (0)

0111 1110 (518466) | more than 3 years ago | (#33623502)

We lowered the wavelength of the laser instead of go to 56x CDs.

How long did it take us to go from a red laser to a blue laser? And speaking of CDs, how much faster are CD-ROM drives now then they were 10 years ago? I still have one of those 72x TrueX Kenwood CD-ROM drives [xbitlabs.com] with multiple lasers reading the disc in parallel. Can you point to a CD-ROM or DVD-ROM or BD-ROM drive that can read my CD-ROMs faster than 72x? How many years has it been since CPUs hit 3 Ghz? The first Core Duo was released almost 4 years ago. How much faster are CPUs now? Technology doesn't always advance smoothly and steadily. Sometimes it stagnates for a while before it can move forward again.

Re:Sure it might hit a wall... (2, Insightful)

glwtta (532858) | more than 3 years ago | (#33623940)

Can you point to a CD-ROM or DVD-ROM or BD-ROM drive that can read my CD-ROMs faster than 72x?

You're right, come to think of it, there haven't been any major advancements in the speed of floppy and ZIP drives, either!

That was the the parent's point: instead of trying to spin CDs faster, we went to DVDs and then BDs (about 4 times faster than a 72X CD).

How much faster are CPUs now?

Considerably. While mostly not relying on increases in clock speed. That was the point.

Density -- SSD ~= 194GB/cc, 3.5" HDD ~= 13 GB/cc (0)

Anonymous Coward | more than 3 years ago | (#33623266)

You can get a 32GB MicroSD card [amazon.com], which has a volume of [wikipedia.org] of (15 mm) * (11 mm) * (1 mm) = 0.165cc, yielding a storage/volume of 32GB/0.165cc ~= 194 GB/cc. A 3TB HDD has (roughly) a volume of [wikipedia.org] 102 mm * 25.4 mm * 3.5 in ~= 230cc. This yields a density of 3TB/230cc ~= 13 GB/cc.

Looks like SSD is roughly 15 times denser than a conventional 3.5" drive. Seems to me it's SSD's scalability that needs work -- not the physical density.

Re:Density -- SSD ~= 194GB/cc, 3.5" HDD ~= 13 GB/c (1)

AHuxley (892839) | more than 3 years ago | (#33623538)

Got to keep cartel pricing up, why race to the bottom just yet :)

Re:Density -- SSD ~= 194GB/cc, 3.5" HDD ~= 13 GB/c (2, Insightful)

ChrisMaple (607946) | more than 3 years ago | (#33624814)

194 GB/cc is about 8e10 atoms per bit, assuming 2 Angstrom atoms. Since it's going to be really difficult to store more than 1 bit per atom, that sets a hard limit of improvement at 8e10 times what's available today.

Slow news day. (3, Insightful)

twidarkling (1537077) | more than 3 years ago | (#33623402)

You know, stories like this used to interest me. Then I noticed that:
a) they kept reoccurring, and
b) had a common theme.

Yeah, it's always "We're approaching a wall with what can be done with current technology, so it's going to either be more expensive, or need a new technique, yadda yadda." Tell you what. Lemme know when we *actually* hit the wall in ANY of these that they keep threatening us with a wall in making, SSD, HDD, CPU size, etc.

Re:Slow news day. (1)

Dadoo (899435) | more than 3 years ago | (#33623968)

Lemme know when we *actually* hit the wall in ANY of these

Well, we've already hit a wall with CPU clock speed, haven't we? According to this [wikipedia.org] page, CPUs hit 3.6GHz in 2005. I haven't seen anything faster, since then - at least, not in the Intel product line. Yes, IBM's POWER chips are up to 5GHz, but that's not much of an increase. If clock speed had maintained its trend, we'd be up to 9 or 10 GHz, by now.

I expect that, in the not-too-distant future (10 years, maybe), the rest of the computing attributes will hit their wall, too.

34nm is better tech than 25nm (4, Interesting)

0111 1110 (518466) | more than 3 years ago | (#33623452)

The smaller the NAND flash process size the shorter the write endurance and data retention times. A 25nm NAND flash SSD will have a much shorter lifespan and hold data for a much shorter period of time than current 34nm tech. Does this mean that 2010 NAND flash SSDs will be better than 2011 ones? Well I guess that depends on how much you value reliability and longevity in your storage devices. Lower cost and shorter life is a win/win for the manufacturers. This limit on NAND flash technology has been known since the start. I don't see the big deal. Just stop at 34nm and work at other technologies that are faster or scale in size better. We usually think of larger process size as being better, but in this case it's not.

http://features.techworld.com/storage/3212075/is-nand-flash-about-to-hit-a-dead-end/?intcmp=ft-hm-m [techworld.com]

http://hardforum.com/showthread.php?t=1492711 [hardforum.com]

Use the 5.25" bays (1)

AHuxley (892839) | more than 3 years ago | (#33623534)

Think of all the desktop cases with space. Layers of chips high and the depth too.
If not, always U2.
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

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>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...