HRDs Faster than SSDs and HDDs

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
I didn't say: "SSD's, why discus this?" before they came out.. or something like that, I just suggest you do not make such a statement like "HDR is total crap", "HDR will never take off", please tell us some facts not just your opinion in 4-5 words
sorry for bad english
 
The HRD only uses 4W of power, a third of the power consumed by HDD, half of the typical SSD.
Click to expand...
Like the other guy said: 8w for an SSD? or even 12w for a HDD? (maybe a SAS on a streaming write...)
But is that 4w constant??? It better be able to drop down to .1w in idle or its trash. Sounds like its expensive to make anyway, with 64 r/w heads
 
oh, and if its the media that moves instead of the heads then the media can only take up 1/4 to 1/2 of the internal lateral area. It won't make for very dense storage... especially since there will be two layers of glass per layer of media.
and how fragile will that drive be? Hopefully heat won't distort the glass or cause it to shatter... maybe they should use Pyrex HAHA but still... I would dare risk dropping one...
 
This almost sounds like fairly land. So if they have 64 read/write heads right now and want millions in this same device later... let's say they can do it, what's stopping them from having a billion? Or just assign a head to each sector of the platter and not even require moving parts. The controller stores a map of the sectors anyway right? This is required to know how to move the head to extract stored data.

What are capacities? Like SSDs or HDs or more?

What would prices be? Like SSDs or HDs or more?

No word on a street date, this could be a few months until the first models or most likely, minimum of two years -_-

Well I was planning on getting two SSDs in October when Windows 7 releases so I don't think I'll be seeing this on the shelf too. Shame.
 
I think it's pretty cool tech. You have a HDD plater that's square since it doesn't have to spin, sandwiched with two layers of microscopic heads. You get rid of the moving heads and spinning drives.

I know it's not exactly like that, but it's a good way to have the image in my head. It's possibly a replacement for future HDD's.
 
[citation][nom]Ciuy[/nom]the ioDrive is an SSD on steroids and btw that costs like what? 15000$ ? i forgot anyway something no1 will buy....theres always something better but its price is outrageous. You can get a 100$ CPU Beat a 500$ CPU with the liquid cooling and stuff but no1 wants that cause its outrageous expensive. io is crap[/citation]
I spoke with the guys over at ioFusion a few weeks ago, inquiring about their prices. There new top of the line ioDuo 640gb drive is about $9600.
 
from an engineer's point of view, it is too costly for such number of read/write heads. even if it become dirt cheap in the future, major problem is one heavy moving part which is pretty unreliable or expensive to be reliable.
we can easily RAID SSDs or HDDs if we wanted to match the capabilities of this type of drive.
 
How much noise does it give out I wonder. Solid state is the way forward.
 
Cool but...

I was wondering what happened to the infinite capacity gelatal memory being developed by the Japanese in the early 90's? I know they were successful getting the data in, but the output was garbled chaos.

This are more moving parts, only the parts that move are larger than the heads, so logic dictates heat at speed, and a higher failure rate. Don’t know about everyone else here but I am looking for increased stability, and only expect minor improvements in speed. With systems like NetApp and others that allow you to put data across many more spinals, we are getting more than satisfactory business productivity speeds now. Could we use faster, yes, but at what point do we spend more time on reliability and less on speed increases? The problem is that reliability does not create return customers looking to upgrade unreliable or (time) antiquated systems, and every honest employee at Seagate, WD, Fujitsu, Toshiba, SIS and others will tell you it is about moving product.
 
With i7 boards holding 12gb of RAM, i thought it would be nice to see an OS boot off system memory. We already have Ramdrive, but we need a motherboard feature to allow a BOOT from RAM.
 
If they would simply create a harddrive platter with 2 or 3 arms per platter (and put the arms in RAID) you'd get the same results.

On the other side of a platter is enough free space to host a second arm,which effectively should double (perhaps even upto 233%) the R/W speeds of a disk.

Hosting 2 arm mechanisms inside a box, and a 500GB HD,should not cost that much more, and the ratio Speed/Power consumption should be less than 2x
 
[citation][nom]hixbot[/nom]With i7 boards holding 12gb of RAM, i thought it would be nice to see an OS boot off system memory. We already have Ramdrive, but we need a motherboard feature to allow a BOOT from RAM.[/citation]
That wouldn't be possible. RAM is erased each time you turn off your computer so you will need to copy the whole OS into RAM each time you boot in order to boot from ram. The copy would take so long it wouldn't be worth it at all!
 
I'd also like to say, this drive doesn't seem that great to me...
Reason: Moving parts...
Any computer technician with years of experience would tell you that MANY errors, blue screen or other are caused by the hard drive (bad sectors, corrupt files etc) and the fact that it is mechanical and WILL BREAK DOWN. (it's just a matter of time)

When talking with my co-workers we all agree that one of the best parts of an SSD is that it has no moving parts. SSD's can be made ultra reliable it doesn't matter if they are right now or when in the future it will happen with no moving parts it's not mechanical and they can be made to last a LONG time!
 
What I do not understand is why static, multiple, full length dual sided reading heads cannot be embedded on a platter in between the data platters (instead of the current solenoid driven arms. This will be equivalent to increasing drive speed by the number of heads on a platter i.e 6 heads at 60 degrees of each other will turn a 5400rpm drive into a 30k rpm monster, and it would be less prone to failure (less moving parts), quieter and lower power consumption.
 
I've heard many, and I'm sure hardcore enthusiasts have heard even more than more of this.
Some random technology, whether it be new or advancing current technology, is going to be so great and blah blah blah...

We still use typical HDDs and CPUs for normal computing don't we?
When was the last honest break through like that in computers? The 3d acceleration card? It's not the SSD because flash memory is also quite matured.


What's the fastest drive?
Either creating a ramdisk on your current socketed ram, or getting a special enclosure with a dedicated UPS housing several sticks or ram fucntioning through your average HDD ports...
 
[citation][nom]hixbot[/nom]With i7 boards holding 12gb of RAM, i thought it would be nice to see an OS boot off system memory. We already have Ramdrive, but we need a motherboard feature to allow a BOOT from RAM.[/citation]
A new dell Precision laptop can utilize 32GB of RAM.

http://www.tomshardware.com/news/precision-mobile-laptop-32gb-quadro,10898.html

and soon all new laptops (2011, 32nmm) will be able to utilize 16GB .. i think.
It is too bad that RAM circuitry is not very compact and so we cannot use it for storage (that, and volatility).
 
[citation][nom]ProDigit80[/nom]If they would simply create a harddrive platter with 2 or 3 arms per platter (and put the arms in RAID) you'd get the same results.On the other side of a platter is enough free space to host a second arm,which effectively should double (perhaps even upto 233%) the R/W speeds of a disk.Hosting 2 arm mechanisms inside a box, and a 500GB HD,should not cost that much more, and the ratio Speed/Power consumption should be less than 2x[/citation]

That would cause the hard drive manufacture to have to put in 2.5" platters instead of 3.5" platters into a 3.5" drive, because the secondary (or more) servo arms take up a good corner chunk of the drive chassis.

Piezoelectric wrists on the servo arm ftw-ish. See 2tb caviar black. i hope they bring this to the scorpio series, add more wrists, and longer strokes with the wrist, and asynchronous movement from platter to platter.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom