Slashdot Log In
640gb PCIe Solid-State Drive Demonstrated
Posted by
Zonk
on Fri Sep 28, 2007 01:04 PM
from the super-fast-super-small dept.
from the super-fast-super-small dept.
Lisandro writes "TG Daily reports that the company Fusion io has presented a massively fast, massively large solid-state flash hard drive on a PCIe card at the Demofall 07 conference in San Diego. Fusion is promising sustained data rates of 800Mb/sec for reading and 600Mb/sec for writing. The company plans to start releasing the cards at 80 GB and will scale to 320 and 640 GB. '[Fusion io's CTO David Flynn] set the benchmark for the worst case scenario by using small 4K blocks and then streaming eight simultaneous 1 GB reads and writes. In that test, the ioDrive clocked in at 100,000 operations per second. "That would have just thrashed a regular hard drive," said Flynn. The company plans on releasing the first cards in December 2007 and will follow up with higher capacity versions later.'"
Related Stories
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
Oblig. (Score:5, Funny)
Re: (Score:3, Funny)
Re:Oblig. (Score:5, Insightful)
Even if you get a 32GB model, you can install windows on it and use the regular SATA2 HDD for movies/music storage. Think of the booting time.
Parent
Re: (Score:3, Interesting)
Why even have swap files? Shouldn't caching decisions be done a bit more intelligently at the application level? I have 10 times more RAM in my current PC than all of the memory (including the HDD) on my first PC. At some point, can't we drop swap?
Re: (Score:3, Informative)
Once you get above $500 desktop computers, it doesn't much matter. A properly tuned system will only use swap, if at all, to drop a few MB from RAM to disk because it's just never accessed. A server that swaps during use is just not set up properly.
Re: (Score:3, Informative)
Re: (Score:3, Insightful)
So we still need
Re:Oblig. (Score:4, Informative)
Hard disks have a fragmentation issue because sequential accesses are much faster than random ones with a spinning disk. Each time the next sector to get isn't right after the previous one, the head must seek to the start of the next track. Solid state "disks" have true random access, where accessing blocks in random order costs no more than sequential accesses. So while solid state will fragment, it doesn't matter for performance or reliability.
Parent
Re: (Score:3, Funny)
Yes, but it *needs* to be defragged because I *hate* seeing red in my disk analysis...
Re: (Score:3, Informative)
AFAIK for flash, sequential access is still faster than random access, even for NOR flash.
It's quite amazing how slow some flash is (esp NAND flash). Some are 1ms for random access, and others are even 7ms.
For comparison a 15krpm drive has a random access time of about 5-6ms. So if you have RAID10, you might get similar or faster speeds, than a flash drive 10x the total price of a RAID
Re:Oblig. (Score:5, Informative)
'worn out' flash doesn't spontaneously change state. Bits just get stuck and don't erase correctly.
I don't know how flash drives actually handle this, but it isn't magic or impossible to fix.
Also, the lifetime of modern flash is long enough that it is hardly an issue any more, even for normal desktop use. Maybe you don't want to use it for swap *IF* you swap a lot, but given the cost is in the same ballpark as RAM, you could just buy more RAM.
Parent
Re: (Score:3, Informative)
Re:Oblig. (Score:5, Interesting)
Explanation:
Most flash vendors have moved to MLC (Multi-Level Cell) flash. It's cheaper and denser, but the bit-error rate goes up because you have more bits per cell. The typical life expectancy for MLC is somewhere in the range of 10,000 writes using single-bit error correction. This is compared to 'older' SLC flash which has a write endurance of 100,000 to 300,000 writes.
Now, most vendors making media out of flash take varying degrees of a combination of two approaches (in addition to standard wear-leveling approaches). The first approach is to assume that the majority of the users will only ever store audio or video data so the occasional uncorrectable error won't have much impact as long as it doesn't corrupt the filesystem. The second approach is to use more advanced error correcting algorithms to compensate for the higher bit-error rate.
Using more advanced algorithms, it's possible to get more than 300,000 writes out of a MLC flash-block before the errors become uncorrectable.
P.S. I may be wrong, but I believe flash can have some really odd error conditions. For example, it's possible to disturb a bit in a block just by reading it. I believe it's also possible to disturb a bit in a different block on the same matrix when writing. That's why some form of error correction is always required with NAND flash.
Parent
Re:Oblig. (Score:4, Informative)
The specs for a 256Mb NAND flash memory chip [alldatasheet.com] by Samsung (which is by far the biggest NAND flash manufacturer today) quotes 100k millon write/erase cycles, and this is for an IC commonly used in USB pendrives. The figure usually tends to get worse with increased memory sizes since the memory "element" (float gate) becomes smaller. For example, Modern 16Mb chips, which are the ones i have experience with, usualy quote 1 million W/E cycles endurance.
But, it felt good stroking my ego a bit more
Parent
Re: (Score:3, Informative)
-nB
Uhh, Price? (Score:5, Interesting)
Who, what, when, where, why?
Price would seem to be a pretty important detail...
Re:Uhh, Price? (Score:5, Funny)
Parent
Re:Uhh, Price? (Score:5, Informative)
Parent
Why the product rocks.... (Score:3, Informative)
If your looking to run a blast/darwin query on 50k files to find the closest match to an unknown dna sequence Either you need to recode a bunch of software to use sql, or you snag a piece of hardware that gives database level performance. 80gigs at $2400.00 is a bloody bargain.
The device is also 10x faster in bandwidth than a normal drive which is comparable to a san, but not such a power hog.
So really the tradeoff rocks for small
Re: (Score:3, Insightful)
Re: (Score:2)
Re: (Score:2)
At $30/GB, that would make an 80GB drive around $2,400. Hopefully the price will eventually come down, as "He even hinted that the company is looking into some gaming applications, but didn't want to give any further details."
Re: (Score:3, Insightful)
Re:Uhh, Price? (Score:5, Informative)
Of course, large back then meant 4G, and the average hard disk was 9G. This is evolutionary, not revolutionary.
Parent
Re: (Score:3, Interesting)
It is. But so is PCI -- it's just a matter of whether the physical and electrical connections on your hardware allow it, and whether your OS is set up to handle it. I wouldn't recommend yanking out your video card while the PC is on.
The return of HSM? (Score:2)
And another question. (Score:5, Insightful)
Re:And another question. (Score:4, Informative)
They appear to want to use normal DRAM memory for the running of the drive but then write it permanently to the NAND flash at shutdown/memory full time.
I would assume this involves charging of a small battery and dumping the data later on.
http://www.theinquirer.net/gb/inquirer/news/2007/09/26/hitachi-reckons-solid-state [theinquirer.net]
Parent
Wow (Score:4, Interesting)
Re:Wow (Score:5, Interesting)
- hardware changes
- hardware initialization (e.g. loading firmware)
- searching for drivers
- applications acquiring and releasing resources and checking for stuff like library versions, user names etc.
That's why BIOS initialization often takes time, and yet it works even if the system has no drives.
The only way this would work is hibernating, but hardware would still need to be initialized.
Parent
Re:Wow (Score:4, Funny)
Parent
Re:Wow (Score:4, Insightful)
Parent
Re:Wow (Score:5, Funny)
It plays very nice.
[turns to the other Commodore users] I told him it plays very nice [chuckling from users]
Now! Go away or I shall taunt you a second time.
Parent
Can you boot from them? (Score:2)
Number of write cycles? (Score:2)
Expect to pay big (Score:2)
on behalf of all of slashdot, i would like to say (Score:5, Funny)
$30 per gb, ouch (Score:3, Insightful)
PCIe (Score:2)
But it doesn't have a PCIe slot. Something like this would finally give me a reason to build an all new PC. Anyone e
Misleading benchmark (Score:5, Interesting)
By which he means, set up a completely unrealistic benchmark which shows his flash drive in the best possible light, and a traditional drive in the worst possible light.
I still want one of these, but that benchmark is nothing to be proud of.
Re: (Score:3, Insightful)
No, the scenario he set up is a classic worst case scenario for drives, one which is well known to large disk usage corporations: near-random access by hundreds-or-thousands of concurrent users. This is what SANs are built to address, this is what parallel RAID
HardCard! (Score:3, Interesting)
Talking to the company at demo (Score:5, Informative)
That being said, a few of the guys there said that they pretty much expect these (at the beginning) to do the best sales for companies that are looking to get really really fast database servers going. NOT for scsi san replacements (it's silly to spend $100,000 for something you could get for 10,000 hard drive space wise). Eventually as the price drops... i know of a handful of people who would EASILY pay 1000$ to get one of these on a gaming rig even if it was only 100 gigs. But that right there is already 1/3rd of the price of what it currently is. (assuming it's around 30$ a gig).
Another thing to keep in mind that came up in the conversations... since these are tiny, think about the cost per server rack... and think of the cost per electricity to run. If you take those into consideration, these are actually less expensive that most people would think! A massive rack of hard drives could cost a lot of money in a co-location ... and a lot of electricity to run it all... But then again, we're talking about savings on servers, not general in home use.
When this gets to about 1/3rd of it's current price, that's when you will see these things become TRUELY mainstream both to the average company and home users (be it rich ones who need the latest and greatest).
Fusioni-io [fusionio.com] -- Link to their site.
Re: (Score:2)
I hope this means that laptops and large capacate media players with extremely long battery life are not too far away.
At least as far as laptops are concerned, I thought a limited number of write cycles was a problem for solid-state drives.
Re:Still Expensive (Score:5, Insightful)
I think people expect too much from SSDs. The hard drive is far from the dominant power consumption component in a notebook. The CPU, chipset, GPU and display panel each consume more power than a notebook hard drive does. If you follow a modified version of Amdahl's law (not a law, but whatever), you want to fix the biggest problem first, and that is either the display or CPU. An LED backlit display can save some power, and running a lower power rating CPU saves power too. Compared to that, the savings of swapping HDD for SSD is negligible. On a standard notebook, I think you might add 15 minutes to battery life, which is still far from "extremely long battery life".
In media players, doubling in capacity every year is a reasonable expectation.
Parent
$30 bucks a gig (Score:3, Informative)
That means their low-end 80GB drive will be around $2400+ or so US dollars depending on tax, shipping, retail prices etc.
Re: (Score:3, Funny)
The fuck?
by egg troll (515396)
Oh, carry on.
Re:write limit? (Score:4, Insightful)
Parent
Re:write limit? (Score:5, Informative)
Cosidering that this drive is 640GB, that means you would need to write somehwere in the region of 61 PETABYTES of information.
You'd have to write to the drive at a perfect 800 MB/s for 941 days to hit that mark.
It could last as long as 30 years, at full write speed of 800 MB/s if it can handle 1M writes per cell.
At the end of the day, semiconductors this large and high quality are certainly better than tiny bits of rust on rapidly spinning platters.
Parent
Re: (Score:3, Informative)