Details of Intel's biggest solid-state drive so far, a 320GB part built on its 34nm process, are popping up across the web. The current X18-M and X25-M models come in 80GB and 160GB capacities, use 2bits per cell multi-level cell (MLC) technology and are built on a 50nm process. The single-level cell (SLC - one bit per cell) …
Disk vendors should invest in defrag tech
I am surprised that the existing hard disk manufacturers aren't investing in and giving out free disk optimisation software in an attempt to increase the longevity of "spinning platter" hard disks. Also they should be working with the AV vendors to stop them hammering the disks so much & ruining battery life on laptops.... and MS to stop XP et al pointlessly paging to disk on a machine with 2Gb _free_ RAM... thus making normal HDs look better & reducing power consumption.
SSDs will become mainstream (i.e. when my mum has one in her laptop) sooner or later but for now most people still have hard disks. OS performance (especially Windows) is crippled by hideously fragmented file systems... even on a basic computer an _intelligent_ optimise can make a huge difference to computer responsiveness. Move archiveable files towards the spindle, sort frequently used files, clump frequently modified files & leave some room for expansion... block placement of directories for DBs with expansion room after certain files... etc. etc.
I think users who consider disk optimisers resent the rip off prices some of the vendors charge (especially as some are very poor). Bearing in mind how much difference it can make to a machine & the perceived speed improvement of the hard disk, I am surprised that the disk vendors don't get involved here! Make your tech look better... sell more kit from your existing assembly lines.
Some disk vendors do SSD also.
1.8" HDD is all but dead.
Even less endurance?
The problem with another die shrink on the underlying NAND is that this further reduces "writes per cell" - with every generation an order of magnitude is lost.
While this is great for the mobile industry, much larger capacities in less space - its fine for iPods and phones where we tend to "write once" and read many - how many times do you delete an mp3 or a photo and then overwrite the blocks ? never?
In an disk drive however we are forever re-writing blocks. If endurance drops to <10,000 write per cell for these 32nm NAND, then you have to over-provision more and more - thus defeating the purpose of adding the extra capacity. Todays 45nm SLC NAND is the only real option for an enterprise drive - I wouldn't even like to put an MLC device in my desktop or laptop until we have hybrid SLC and MLC - so the disk itself can move stagnant data onto the MLC portion.
RE: Even less endurance? #
MLC may understandably loose some endurance with smaller process geometries, but come on, an order of magnitude? 10,000 writes per cell goes to 1000? Hardly...
Also, MLC lifetimes on these drives have been calculated a million times, and the numbers I've seen are always ridiculously high in endurance e.g. you can write over the entire capacity of the drive every single day for 5+ years without issue.
Eventually, I'm sure the technology will have to change to new forms of memory storage, but even before then I'm sure SLC will just become the standard as dies continue to shrink.
RE: Even less endurance?
A couple of generations back, SLC was at over 1,000,000 raw writes per cell. Now we are at 100,000 - so yes there is a direct correspondance with the number of electrons per bit and how stable these are over time.
Maybe its not down to 1,000 but it will be lower than 10,000. With MLC drives today, performance is a lot lower than SLC - thus the number of writes you can do in 5 years reduces - however, the only reason that 'effective' lifetime can be 5 years is because of the over provisioning of the raw capacities - hence my comment re the need to over provision more. Look at some of the latest MLC devices - these have 50% effective capacity - vs much higher effective capacity on SLC devices.
I think we are in agreement however, SLC will become the standard requirement as dies shrink for the SSDs used to replace HDD, but MLC will continue to be driven by the consumer (mobile) requirements.
HD Manufacturers aren't lost.
Price of an SSD scales linearly with capacity. Price of an HD doesn't. It'll be quite a while before one can get a Terabyte of solid-state storage for the cost of an HD.
And I'm not optimistic that Moores law will last very much longer. You can't make smaller atoms. For that reason I don't anticipate Terabyte chips, ever.
BTW the best solid-state storage performance would not attach to a disk interface, it would be a card that plugs in to a PCI-X slot.. That's also a more natural manufacturing format for a big bunch of chips. And if HD manufacturers get optical write addressing working, the physical limits are hundreds of Terabytes per disk. (What would one do with a hundred Terabytes inside one's PC?)
My guess is that a future PC will have maybe 40-80Gb of solid-state storage soldered onto the motherboard and a hard disk as an option when the user needs larger amounts of local storage. A smart way to use the solid-state storage would be as persistent cache for a network filesystem. The problem with diskless workstations has always been that they forget everything when they are powered down and then thrash the network and the server when they power up.
capacity not speed
The disk is used for capacity not speed; cache is used for speed not capacity. Many application files once they exceed MB they slow down the processor. But even the largest streamed video file will not require more than 1 GB, so I think the high-speed performance aspects of drives are overkill.