The World's Fastest Hard Drive?

LazyLion

LazyLion

King of de Jungle
Joined
Mar 17, 2005
Messages
107,579
Reaction score
10,221
Location
District 9
http://www.geeky-gadgets.com/?p=12225

The World’s Fastest Hard Drive - Fusion ioDrive Duo
Posted 12 Mar. 2009 in Gadgets, PC Hardware by Fatgadget

Fusion has added a new drive to their range of super fast, extremely expensive hard drives, with the launch of the new Fusion ioDrive Duo.

The Fusion ioDrive Duo is according to the manufacturer, the fastest SSD solution available today, and from the looks of the specification they might be right.

The ioDrive duo comes with 160GB of storage, for the basic model, with up to 1.28TB for the top model, it has read speeds of 1.5GB per second and write speeds of 1.4GB per second.

It is designed for the server and enterprise market and can be used in a RAID-1 setup for maximum performance..

It is possible to use the RAID-1 setup with just one card, between two memory modules, as the other ioDrives it connect to your PC via a PCie slot.

There is no word on pricing as yet, it is due to go on sale in April, I bet it is going to cost a fair bit when it is launched, especially the 1.28TB model.
Click to expand...

fusion-iodrive-duo.jpg


I'll take two... thanks! :D
 
this thread is useless without that kinda money.
 
you mean companies.

this isn't meant for home use.....yet.

there isn't even a price yet but i'm guessing anywhere between R50 000 to R500 000
 
What is the seek time of this drive?

No pointing having a drive that can transfer fast but access slow. Not for your home user anyhow :)
 
Since its NAND memory based probably sub 1ms - it's been reported that 320GB will cost somewhere around $14,000.00 which means you'd probably need to be a corporate.
 
Just wondering i am not sure if the the way it connects would slow it down.

So your saying it has the same seek time as other ssd drives even though it connects via a different interface?

Unfortunately, the ioDrive Duo does come with some asterisks. It cannot be used as a boot drive, and due to its drivers also requires the use of a 64-bit operating system. While the price is TBA, the original 80 GB ioDrive is still over $3,000, meaning the 160 GB ioDrive Duo will be at least that expensive, with the 1.28 TB version probably going for five figures. Whether you plan to spring for one or not, the ioDrive Duo has certainly set a new standard for SSD speed on a myriad of levels. With Steve Wozniak now on the Fusion-io payroll, who knows what the company will think of next.

Wonder why it cannot be used as a boot drive?
 
Talking about SSDs on servers, it's worth learning about the importance of wear levelling in the OS (or "leveling" as the Americans spell it). Without wear levelling, you can reach EOL in an SSD very much sooner, because the chips support only so many write cycles (currently around a million or so). Win7 is SSD aware and incorporates a high-performance wear leveller.
 
killadoob said:
Just wondering i am not sure if the the way it connects would slow it down.

So your saying it has the same seek time as other ssd drives even though it connects via a different interface?

Unfortunately, the ioDrive Duo does come with some asterisks. It cannot be used as a boot drive, and due to its drivers also requires the use of a 64-bit operating system. While the price is TBA, the original 80 GB ioDrive is still over $3,000, meaning the 160 GB ioDrive Duo will be at least that expensive, with the 1.28 TB version probably going for five figures. Whether you plan to spring for one or not, the ioDrive Duo has certainly set a new standard for SSD speed on a myriad of levels. With Steve Wozniak now on the Fusion-io payroll, who knows what the company will think of next.

Wonder why it cannot be used as a boot drive?
Click to expand...

Perhaps its drivers cant be loaded at boot time - like a RAM disk??
 
Ok arthur bud, lets put that in english for the non greek speaking people like myself haha.

Are you saying some os's will wear the hdd's out in 3-4 years while os's like windows 7 go easy on it? yea could be conrad.

There was a program that you could make your own drive like that but obviously there are limitation with the current memory setup :).

http://www.superspeed.com/desktop/ramdisk.php something like this but i would imagine you need a server board or something and considering how expensive memory is it would be kuk expensive but try and make a drive and copy a file and see how it flies. Make a 1 gig drive and copy say a 800 meg file.
 
Last edited:
Right, you need a 64-bit OS because this hard drive plugs into a PCI-E slot. It pics up similar to video memory, in that 32-bit just doesn't have large enough registers for addressing all the space, unless you're willing to settle for <4gb of hard drive space. That being said, the manufacturer probably didn't even bother making 32-bit drivers either, as a result.
You can't boot off of the device because you need drivers to be loaded first, telling the OS how to talk to the actual hardware, since it has a PCI-E interface. Technically, Microsoft could modify Windows to provide for this, but it would hardly be worth the effort to them.

The idea with these drives, is to have your OS run off of a raptor or some such, and then you keep your databases and other high-load software on these badboys.

As for seek times, you're looking at <1ms. If memory serves, the full press release actually stated 0ms.
 
Messugga said:
You can't boot off of the device because you need drivers to be loaded first, telling the OS how to talk to the actual hardware, since it has a PCI-E interface.

Technically, Microsoft could modify Windows to provide for this, but it would hardly be worth the effort to them./QUOTE]

The device is essentiall a RAW device with no MBR that a current legacy BIOS knows about. EFI might be able to address this. I wish EFI was present on all newer motherboards.

I don't understand how microsoft could write an OS driver to solve this seeing that the bios wont recognise the device in the first place.

The PCI-E part is not the hard one if you think of it in terms of RAID controllers that contain their own bios to boot the os.
Click to expand...
 
Wear levelling is also implemented on the drive itself.

Write endurance rating:- 2 million cycles. (The typical range today for flash SSDs is from 1 to 5 million. The technology trend has been for this to get better.

When this article was published, in March 2007, many readers pointed out the apparent discrepancy between the endurance ratings quoted by most flash chipmakers and those quoted by high-reliability SSD makers - using the same chips.

In many emails I explained that such endurance ratings could be sample tested and batches selected or rejected from devices which were nominally guaranteed for only 100,000 cycles.

In such filtered batches typically 3% of blocks in a flash SSD might only last 100,000 cycles - but over 90% would last 1 million cycles. The difference was managed internally by the controller using a combination of over-provisioning and bad block management.

Even if you don't do incoming inspection and testing / rejection of flash chips over 90% of memory in large arrays can have endurance which is 5x better than the minimum quoted figure.

Since publishing this article, many oems - including Micron - have found the market demand big enough to offer "high endurance" flash as standard products.)

AMD marketed "million cycle flash" as early as 1998.
Sustained write speed:- 80M bytes / sec (That's the fastest for a flash SSD available today and assumes that the data is being written in big DMA blocks.)
capacity:- 64G bytes - that's about an entry level size. (The bigger the capacity - the longer the operating life - in the write endurance context.)

Today single flash SSDs are available with 160G capacity in 2.5" form factor from Adtron and 155G in a 3.5" form factor from BiTMICRO Networks.

Looking ahead to Q108 - 2.5" SSDs will be available upto 412GB from BiTMICRO. And STEC will be shipping 512GB 3.5" SSDs.
To get that very high speed the process will have to write big blocks (which also simplifies the calculation).

We assume perfect wear leveling which means we need to fill the disk 2 million times to get to the write endurance limit.

2 million (write endurance) x 64G (capacity) divided by 80M bytes / sec gives the endurance limited life in seconds.

That's a meaningless number - which needs to be divided by seconds in an hour, hours in a day etc etc to give...

The end result is 51 years!

But you can see how just a handful of years ago - when write endurance was 20x less than it is today - and disk capacities were smaller.
Click to expand...

http://www.storagesearch.com/ssdmyths-endurance.html
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X