SSDs Have Bleak Future, Says Researchers

[Guest]

ssd is doomed?? well yes it is!

you may ask why i second the crazy graduate, and the answer is:
we will use scd´s in the future (solid crystal drives).

you want to know where i get this knowlegde from?
the answer is simple and convincing like the article: i´m a fortune teller at the local circus..!

omfg why did i even bother to read that crap..?

 

[BulkZerker]

[citation][nom]CaedenV[/nom]@TomFREAKThe reason 10K drives do not do well in the consumer market is not due to the performance or price increase, it is due to the noise. In a world where your 7200RPM drive is the loudest part of most computers, a 10K drive just screams. If they managed to make a quiet 10K drive people would be all over them.[/citation]

[Citation][nom]Silentpcreview.com[/nom]At idle, the new Raptor was quieter than every other 3.5" drive in the lab except for our old standby, the Barracuda IV. That includes the older Raptor (only slightly quieter), Samsung's 80 GB Spinpoint models, and Western Digital's own 500 GB Caviar SE16. Deciding between the Raptor and the Barracuda IV was very difficult and it took a several A/B comparisons and a number of second opinions before we finally settled on the Barracuda as the quieter drive. Much of this difference had to do with the subjective quality of the noise; the Barracuda had a more broadband sound that was more muted than the Raptor. That said, the difference was very difficult to pick out from a distance of one meter, even when we knew what to listen for.[/citation]

http://www.silentpcreview.com/article622-page2.html

Now, they took that drive out of it's "cooling cradle" and then put it into a suspension basket. Every problem with noise they had was then GONE, it was QUIETER than a 5200RPM drive. I did what they tested and I can verify how quiet my Raptor is. Whats even funnier, is removing that aluminum basket with all it's "heatsinks" made the drive run cooler as well!

 

[jasonkaler]

Yes, in 2024, 2012 technology will be crap.
I'm pretty sure by then we'll find ways to improve on how it's currently done.

In my opinion, DVDs have a bleak future too, but hey, I'm no expert researcher!

 

[Guest]

Let's use pencil and paper. They never doom.

 

[v1zzle]

2024, really, cause a grad "expert" student can predict what technology will be used in that time..pfff, c'mon. I wonder if she got a lot of props for that study or a more negative approach.

 

[Guest]

Forget SSDs. PROCESSORS as we know them wont be able to keep developing at their pace current after more than 10 years or so. The best estiments from intel are that after a decade we will have run out of ways to stretch that technology (it's getting hard now). If the very heart of all modern computers has to be reinvented or replaced by 2022ish, somehow I think the SSD discussion isn't even worth having.

 

[Guest]

This is like saying

"Cars are going to get bigger and heavier, so using the same 4 cylinder 80HP engine, they will get slower to the point were the horse drawn buggy will surpass the gas powered automobile."

 

[Guest]

Some idiotic grad student comes and splurts out some grand statement and feels important?

Attention whore much?

So basically any idiot can come and spit out big words to get on the title page?

 

[Alex Atkin UK]

[citation][nom]Tomfreak[/nom]Sometimes I wonder, desktop HDD are not limited to power consumption, but why arent the HDD manufacturer push 10K RPM HDD down to mainstream to "slow the SSD" adoption . Surely the latency gap would have improved significantly, 8.9ms vs 4.0 seek time. It may not reach the SSD ones, but it is still better to delay the SSD getting into mainstream. While many of us like the capacity of HDD, but we I would prefer having slighly smaller 500GB with boost speed over the 2TB ones. The size of the mainstream SSD are still less than 64GB, win7 64bit takes up almost half of it, so the leave not much room for games/other thing.[/citation]

Maybe because 10k RPM creates more heat, uses more power and makes it much easier to get a head crash if you accidentally knock your PC (lost count how often and I specifically got an SSD for my laptop because I carry it around while turned on). Oh yeah and 4ms access time? The latest 10K drive on this very site benchmarks at just under 7ms, a far cry from an SSD.

Then of course there is the noise. One big reason I have been switching to SSDs is to get rid of that annoying vibrating noise that burrows into my brain as it rattles even expensive PC cases and even through walls, floorboards, and that is just 5400RPM drives.

HDDs have their place, mass storage for things like video editing, but in those sizes the heat produced with a big RAID means better cooling which will drown out the HDD noise anyway.

So yeah, HDDs likely will remain for some time, but so will SSDs and they WILL get better as there are so many other memory technologies being looked into, likely a better one than NAND will be in production by 2024.

 

[Guest]

I remember back in the early 80's there was an electronics engineer named Steve Gibson who wrote a weekly column for InfoWorld. In one article he predicted that CPUs would never exceed a speed of 25MHz because that would create too much unshielded RFI. The main problem with SSDs that already exists, is that they wear out. Sure, each bit can supposedly survive 10,000 writes, but what if that particular region is used as a temporary cache or for some other function whose contents change regularly? But 12 years from now is an eon in terms of computer technology (12 years AGO we were using Win 98, and a large HD was 600MB), and I expect something much better will have replaced SSDs by that time, perhaps something based on organic components.

 

[Guest]

10K RPM drives are obsolete. Try 15K RPM Fiber Channel (using SCSI3 over Fiber Channel). Loud? Try 100 TB worth of 146GB FC drives.

Now, why would you go for that sort of thing when for the same money as a modern 340GB SAS or FC 15K RPM drive you can get a 480GB SSD? Long term reliability I think. Can't be performance. It's suckers. Even modern stuff like Pillar Axiom is built on spinning drives with some concession to the ability to connect SATA-3 SSD as an add-on option for extra cost.

 

[Guest]

Oh no, SSDs wont work in 2024! The horror.
We will have to find a way to get by with petabyte sized hard drives the size of a dime directly connected to our brainsstem. How will we ever manage.

 

[Guest]

Yes, and 15 years ago they probably said "platter based drives probably won't go much past 4TB"

which is why now flash is popular

and once that has run its course, it wont be NAND any more. Hynix is already working on memristor based architectures.

 

[Guest]

I'm too lazy to read all the comments above, but did anyone point out the (to me obvious) point that manufacturers could build SSDs with redundancy to correct for errors as die sizes shrink? It might mean that the rapidly dropping cost curve will start to flatten out at some point.

 

[kartu]

[citation][nom]rex86[/nom]I recently bought myself and SSD for my laptop and after couple of days I would never go back to HDD. NEVER![/citation]
Come back in about 12 month. When it wears off.
Oh, and don't expect some "silly old" bad sector, that could even be read by repeatedly reading the same part, when that thing breaks it will be like exploded hard drive.

 

[Guest]

World will end before we reach 2024
Fallout going to happen
And I don’t think someone survived the radiation is going to search for SSD rather than food and water

Well we still have enough time for SSD

 

[jonjonjon]

are these guys on the payroll of companys that make mechanical hdd?

quantum computing FTW!

 

[jonjonjon]

are these guys on the payroll of companys that make mechanical hdd? quantum computing FTW!

 

[skgiven]

The obvious solution is to use RAM, properly; basically load everything into RAM and run everything from RAM. That way you read and write to the disk when booting and periodically to backup work. Faster, safer, lasts longer... The technology is already out there, it's just not being used properly (by operating systems).

 

[ipwn3r456]

[citation][nom]skgiven[/nom]The obvious solution is to use RAM, properly; basically load everything into RAM and run everything from RAM. That way you read and write to the disk when booting and periodically to backup work. Faster, safer, lasts longer... The technology is already out there, it's just not being used properly (by operating systems).[/citation]

RAM is a good solution, however, the HUGE disadvantage is that RAM needs to be constantly powered, which means, if there's a power failure, ALL of your data is GONE.

 

[Guest]

Grad students need research to graduate, and publicity to get a job. Well done.

 

[Guest]

This article fails to consider that MRAM (Magnetoresistive storage) will overtake NAND flash by then. It'll be a different ballgame.

 

[mazty]

[citation][nom]olaf[/nom]Those that can do, those that can't teach.[/citation]
Those that can, research. Clearly you've no idea about how research works. Go back to Wendy's.

 

[einheriar]

[citation][nom]stingstang[/nom]" menufactoring process .[/citation]
Just wondering is that the process where one puts prices on dishes in a restaurant menu? ;-) Sorry couldn't control myself :-D

 

[Guest]

*cough* MRAM *cough*