Worlds first 6TB HDD, with helium!

[Useful Idiot]

wow, this thing is filled with helium and they tested it at TT.

http://www.tweaktown.com/blogs/Paul...d-in-the-lab-and-a-few-quick-tests/index.html

looks pretty cool!

 

[Devon68]

Yeah seen it before. It's nice that they are making improvement's in disk capacity. But looking at statistics bigger drives fail faster.

 

[BUCK NASTY]

I'd hate to see my complete HD movie library wiped out by a single drive. I'm pushing my luck with the 3TB drive i just installed in the HTPC. LOL

 

[Useful Idiot]

You guys are wet blankets.

 

[FordGT90Concept]

The read/write speed on it is pretty disappointing to me. That drive would take half a day to format.

 

[Useful Idiot]

It actually takes a few nanoseconds.

 

[FX-GMC]

The read/write speed on it is pretty disappointing to me. That drive would take half a day to format.

What does formatting entail to you?

 

[15th Warlock]

The read/write speed on it is pretty disappointing to me. That drive would take half a day to format.

Quickformat?

 

[FordGT90Concept]

What does formatting entail to you?

Checking for bad sectors. Format without checking for bad sectors is moot.

6,000,000 MB / 150 MB/s = 40,000 seconds = 666.67 minutes = 11.11 hours

DoD 5220.22M 3-pass erase + 40% verification is expected to take over a day and a half (~37 hours) to complete. That's ridiculously long.

3 TB drives take about 4-6 hours compared to 11.11 hours. What I'm getting at is that, even though these drives were able to double their capacity, they were not able to double (or increase at all) their read/write performance. If that is the trend with helium drives, helium drives are not a long term answer to the hard drive problem.

 

[FX-GMC]

Checking for bad sectors. Format without checking for bad sectors is moot.

6,000,000 MB / 150 MB/s = 40,000 seconds = 666.67 minutes = 11.11 hours

DoD 5220.22M 3-pass erase + 40% verification is expected to take over a day and a half (~37 hours) to complete. That's ridiculously long.

3 TB drives take about 4-6 hours compared to 11.11 hours. What I'm getting at is that, even though these drives were able to double their capacity, they were not able to double (or increase at all) their read/write performance. If that is the trend with helium drives, helium drives are not a long term answer to the hard drive problem.

Valid point. Knock on wood that not checking for bad sectors on a new drive hasn't bitten me in the ass.

 

[The Von Matrices]

Yeah seen it before. It's nice that they are making improvement's in disk capacity. But looking at statistics bigger drives fail faster.

He filled HDDs are nice, but in my view it's a dead-end technology. He-filled hard drives are a one-time doubling in capacity. I don't see how they can improve on the technology from here. The form factors are fixed so you can't increased the physical size of the drive, and if someone knew how to make thinner platters with the requisite rigidity (in order to increase the number of platters), it would not be exclusive to He-filled HDDs.

In contrast, the true advances in HDD technology are still platter density. If HAMR or patterned media actually ever comes to fruition then it will be good for improvements over many generations just as has occurred with PMR, which increased capacity by ~10x over many years.

3 TB drives take about 4-6 hours compared to 11.11 hours. What I'm getting at is that, even though these drives were able to double their capacity, they were not able to double (or increase at all) their read/write performance. If that is the trend with helium drives, helium drives are not a long term answer to the hard drive problem.

And therein lies the problem with He HDDs. Since HDDs only read from one platter at a time and all He HDDs do is increase the number of platters, transfer rate does not increase in a He HDD. Therefore the time to write the entire drive increases proportionately to the capacity increase.

 

[theonedub]

At least it's not as bad as the recent 128GB MicroSD cards that take 3 days to write the entire capacity.

How did they come up with 3 days? I've been able to write the full capacity to my 128gb MicroSD in about 3 hours

 

[FordGT90Concept]

And therein lies the problem with He HDDs. Since HDDs only read from one platter at a time and all He HDDs do is increase the number of platters, transfer rate does not increase in a He HDD. Therefore the time to write the entire drive increases proportionately to the capacity increase.

Increased density + same RPM (and it is the same at 7200 RPM) should translate to increased read/write performance because the head is exposed to more data. This is why single platter 1 TB drives are so much faster than dual platter 1 TB drives. I'm beginning to get the impression that they didn't increase the density at all--instead, they used helium to add more platters. But no, they can't be right either because 3.5" drives only have room for 5 platters...unless helium enabled them to make the platters thinner so there's six 1 TB platters in there. That would explain the (poor) performance.

In any case, this is disappointing. I hope it is just a bug or fluke that will get corrected in mature models.

 

[xorbe]

It just sounds like a ticking time bomb to me, it's gonna leak out with time.

 

[The Von Matrices]

Increased density + same RPM (and it is the same at 7200 RPM) should translate to increased read/write performance because the head is exposed to more data. This is why single platter 1 TB drives are so much faster than dual platter 1 TB drives. I'm beginning to get the impression that they didn't increase the density at all--instead, they used helium to add more platters. But no, they can't be right either because 3.5" drives only have room for 5 platters...unless helium enabled them to make the platters thinner so there's six 1 TB platters in there. That would explain the (poor) performance.

In any case, this is disappointing. I hope it is just a bug or fluke that will get corrected in mature models.

I think you had unrealistic expectations, because the helium has nothing to do with advanced platter designs or increases in areal density. The only point of helium HDDs is that the helium allows a reduction the head fly height and thus the ability to cram up to 7 platters into the disk instead of the maximum 5 platters in a conventional air HDD. That's where 100% of the capacity increase comes from. There is nothing else new about them. That's why I see helium HDDs as less than a disruptive technology since it's a one time capacity increase. Air filled HDDs will always coexist with Helium HDDs; the Helium HDDs will just be the new top of the line. They will stay at about 50% more capacity than the top of the line air filled hard disk but no better performance compared to air filled HDDs since they will both be based on the same platter technology.

How did they come up with 3 days? I've been able to write the full capacity to my 128gb MicroSD in about 3 hours

You're right - I stated the wrong unit. 128000MB/14.6MB/s = 2.46 hours

 

[Aquinus]

Increased density + same RPM (and it is the same at 7200 RPM) should translate to increased read/write performance because the head is exposed to more data. This is why single platter 1 TB drives are so much faster than dual platter 1 TB drives. I'm beginning to get the impression that they didn't increase the density at all--instead, they used helium to add more platters. But no, they can't be right either because 3.5" drives only have room for 5 platters...unless helium enabled them to make the platters thinner so there's six 1 TB platters in there. That would explain the (poor) performance.

In any case, this is disappointing. I hope it is just a bug or fluke that will get corrected in mature models.

I think that it does. The helium is supposed to reduce air friction inside the drive which would make it easier for the motor to spin more platters and because of the helium there is less air turbulence than traditional drives. Personally, I would rather see higher density platters.

 

[FordGT90Concept]

In which case, I wonder what their pricing is going to be. Specifically, will two 3 TB normal drives be cheaper than one 6 TB helium drive? If the helium drives are substantially cheaper then I could see them getting used in file storage scenarios where speed doesn't really matter.