Slashdot Mirror
NAND Flash Density Surpasses HDDs', But Price Is Still a Sticking Point (computerworld.com)
Lucas123 writes: With the introduction of 3D or stacked NAND flash memory, non-volatile memory has for the first time surpassed that of hard disk drives in density. This year, Micron revealed it had demonstrated areal densities in its laboratories of up to 2.77 terabits per square inch (Tbpsi) for its 3D NAND. That compares with the densest HDDs of about 1.3Tbpsi. While NAND flash may have surpassed hard drives in density, it doesn't mean the medium has reached price parity with HDDs — nor will it anytime soon. One roadblock to price parity is the cost of revamping existing or building new 3D NAND fabrication plant, which far exceeds that of hard drive manufacturing facilities, according to market research firm Coughlin Associates. HDD makers are also preparing to launch even denser products using technologies such as heat assisted magnetic recording.
Have SSD's reached a point where they have a lifespan comparable to HDD's in the most extreme applications, though? For instance: Just had to replace the HDD in my DVR. It's dual tuner so it's buffering 30 minutes for each channel, perpetually. The HDD lasted for years; would a current-technology SSD last as long before it ran out of write cycles in the flash memory?
Are YOU using the TOOL, or is the TOOL using YOU? Think about it!
If you look at a list of new computers, you will notice that a surprisingly large amount of PCs are already shipping with 128 GB or 256 GB SSD. That's gonna hold everything that most people need. People with bit more specialized needs (hardcore gaming, media production, virtual machines, etc.) can probably soon acquire 1 TB SSD for a price like $200. Only massive data centers will remain as users of HDDs. Flash memory companies are putting huge investments in developing the technology further, while HAMR is still a prototype in skunkworks that is struggling to be usable for mass production.
That isn't how wear leveling algorithm work. Yes, once you hit 99%, every write does involve a rewrite somewhere, but those writes are not concentrated in the 1% free area. Instead, the drive controller is reading sections of already written disk and moving them around.
That's 4.294 Gb/mm^2 and 2.02 Gb/mm^2, respectively, for us SI folks.
But the fact that SSD has caught up HDD quite so quickly means the writing is on the wall.
Quite what is the factor that will keep people buying HDD? At the moment, it's only capacity. With matching densities, matching capacities won't be far off. I've said for the last few years the storage companies should give up on making HDDs or at least plan that way.
You can get a 1Tb 2.5" SSD for a decent price now. And desktop ranges are easily catered for with SSDs and even being supplied by default. The max size hard drive you would really see? It's only 2-4Tb. I don't think it will be "several years", given that you can match capacities now (just by putting multiple 2.5 boards into a 3.5" drive), and the price per Tb is dropping fast, while HDDs are offering nothing over SSDs any more.
Sure, the top-end brand-names will be behind everyone else as they ensure reliability, but it will only be a couple of years before people are basically ignoring HDDs in purchasing.
I don't like the idea of someone trying to fix my HDD with a HAMR.
Minimum threshold fixed. Thanks!
Most newer SSDs are designed to fail gracefully. When they die, they become a read-only device. All your data is still accessible. Many USB flash drives are designed to fail the same way - if you've ever had a USB flash drive mysteriously become "write-protected", it probably died and set itself to read-only mode. Unfortunately, Samsung seems to be one of the SSD manufacturers which hasn't yet adopted this philosophy for failure. But I can understand their reasoning because...
That problem was solved in the 2000s with wear-leveling algorithms. Basically, the "sectors" the SSD presents to the computer aren't actual physical locations. They're virtual locations stored in a table. If the SSD senses certain blocks being used too much or other blocks sitting unused, it moves the data around behind the scenes so that writes hit all flash memory cells about equally. It updates the virtual table every time it does this, to fool the computer into thinking the drive is physically the same as it has always been.
The rated endurance on most consumer SSDs is around 2000-3000 cycles. For a 250 GB SSD, that means you can write 625 terabytes to it before expecting a failure. If you write 100 GB of data to the drive every day, you can expect it to last nearly 20 years. In torture tests, most SSDs have lasted about 2-3x longer than their rating. And no, the first cell failure is not catastrophic. Pretty much all SSDs have a number of reserve cells sitting on the sidelines to take over for cells which fail early.
If your duty cycle is higher than 100 GB/day, they make special enterprise SSDs rated for 10k-100k writes per cell. The price is correspondingly higher of course, primarily due to using SLC (one bit stored per cell) instead of MLC (2 bits) or TLC (3 bits).
Limited number of writes were more of a problem in the early days of SSDs when they were like 32 GB in size. In that case, the exact same characteristics as the above 250 GB SSD would yield only 2.2 years of longevity. But the problem has pretty much become a non-factor as capacities have increased.
Fill up your SSD to 99% and it usually has between 20 and 40% free space to work with (more for enterprise drives, less for cheap drives). Oh wait, you've never heard of over-provisioning?
Hint: every SSD has *at least* 6% extra space for wear leveling - 1TB drives are internally 1024TiB.
P.S. If you wanna counter my first argument, fill your SSD up to 99% and then try to work with it continuously for quite some time. That 1% will get overwritten multiple times and your whole SSD will be prone to a failure.
P.S. Bullshit
SMART Attributes Data Structure revision number: 1
Vendor Specific SMART Attributes with Thresholds:
9 Power_On_Hours 0x0032 094 094 000 Old_age Always - 28138
177 Wear_Leveling_Count 0x0013 097 097 000 Pre-fail Always - 98
241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 9528109928
That's a 100% full 128GB Samsung 830 - there's a headerless dm-crypt volume on it, so from the point of view of the drive every single user visible block contains data. ... 100 years or so.
4.87TB written and it's at 98/3000 erase cycles, a WA factor of about 2.76.
Considering that's after 3 years of continuous operation, at this rate it should hit the rated erase count in
Yeah, I had an OCZ Vertex II that I sent back at least 3 times for replacement. It gave me such a bad impression of SSDs that I didn't get another one for a long time.
My current Samsung SSDs, however, have been running flawlessly for 3 years now.
My eyes reflect the stars and a smile lights up my face.
It should be noted that while SLC flash is good for around 100,000 writes or so, TLC flash is only good for around 1,000. MLC is in in-between, about 30,000 writes. So the type of flash used in the drive very much matters.
Even Capacity is going to go this year, with expected 16 TB SSD drives coming to market. The only thing Spinning drives have at this point is Price. And if price is all you care about, then go Cheap! For everything else, go SSD>
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
Some swapping of live data occurs, but having extra slack free space to move around in helps the algorithm better work within those constraints. In fact, Samsung provides a utility called Magician to manage Over Provisioning for extended life. It's not required, but ostensibly it does help.
Life is not for the lazy.
The OP rather implies that a supplier offering both convetional HDDs and SSDs of the same capacity would offer their products at prices based upon "cost of manufacture + margin" - i.e. that the retail prices would be a reflection of production costs. Sadly for consumers, this is blatantly not the case. The evidence for this is *everywhere* - for example a BluRay Movie costs no more to make, ship and sell than a DVD [maybe less, the packaging is smaller, lighter and cheaper to ship] and yet BluRay discs cost significantly more. Another classic example is the motor trade, where 2 cars that are identical in every respect except the engine size are priced so that the one with the larger engine costs more. Going back to the storage industry, there may be at least a couple of legitimate reasons for the price differential : first, the vendor is still recouping research and development costs from SSD technologies, whilst HDDs may be investing much less in R&D and therefore cost less. Second, economies of scale mean that a vendor can spread overheads across greater sale volumes and thus one format costs less. Unfortunately, what is most likely to be happening is that vendors are "fixing" market prices and using the principle of "cool new thing" to charge a premium for the latest product, well beyond what legitimate development costs would suggest. In theory many countries have national agencies to stop markets conspiring to fix prices like this. There is legislation against this [it's essentially racketeering and/or market manipulation, after all]. Unfortunately, 99% of the time, large suppliers get away with it. It's only when something goes unexpectedly wrong [look at the LIBOR rate-rigging scandal in the UK] that regulators will act [because it puts them in a position where they have no choice but to act]. Unfortunately, for the rest of us, for most of the time, a price is set on the basis of "the maximum we can get away with", as determined by the vendor.
Yes, the SSD does have a separate tracking algorithm to manage dynamic LBA mapping to cells for wear-leveling. And yes, and abrupt power outage can corrupt and brick the drive. The OCZ Vertex series have a history of this happening where it can't decrypt (internal) and mount the value due to said corruption. Newer SSDs such as the prosumer and enterprise variety include extra capacitance to ensure half-writes don't occur and thus recover from both a firmware and OS journaling file system error.
Life is not for the lazy.