Helium-filled drives announced

Helium-filled drives announced

Summary: WDC's HGST unity announced helium-filled drives that ship next year. Great technology with a real bottom line benefit.

TOPICS: Storage

HGST - formerly Hitachi's disk unit - announced a major advance in disk capacity and efficiency with their development of helium-filled disk drives. The drives enable greater density in both platters and bits, run cooler and are more energy efficient.

It's a major win for disk technology.

The numbers

HGST says that the new, helium-filled sealed-drive platform reduces power consumption by 23% compared to an equivalent air-filled drive side-by-side. It also enables them to add 2 additional platters - for a total of 7 - to their 3.5" disks, which should mean ≈6TB drives.

The drives also run 4°C (7°F) cooler. They calculate that the total improvement in watts-per-TB is 45 percent, which they expect to improve as the helium-filled drive platform enables higher drive capacities in future products.

The science

Drive engineers have long known that the air inside drives creates turbulence and drag that degrades performance and efficiency. The problem has been what to do about it.

Drives today are not hermetically sealed: they have a filtered air channel to allow pressure equalization between the outside and inside of a drive. HGST engineers have evidently figured out 3 things:

  1. How build a drive structure stiff enough not to flex with pressure changes.
  2. How to seal in helium, a gas known for its remarkable diffusion properties.
  3. How to manufacture such drives in high volume at - one hopes - a competitive price.

The only things better than helium in a drive are hydrogen - which is flammable - or a vacuum - which would need a stiffer, heavier and more expensive frame.

They said it took 6 years of development. I believe it.

The Storage Bits take

While we're waiting for the drive engineers to perfect heat-assisted magnetic recording (HAMR) and patterned media - key technologies for the next major jump in recording density - helium-filled drives will enable higher density and larger capacity drives for years to come.

The major buyers will likely be large cloud storage providers, where a 45% increase in energy efficiency per TB will pay major dividends in power and cooling costs. But expect the technology to spread to all drives over time as it is perfected.

There will be a price to pay: helium is more expensive than air; and a stiffer, diffusion-proof disk structure will also have costs. But with volume both costs should drop over time.

It's great to see this kind of basic R&D come to fruition. Congratulations to the HGST engineering teams.

Comments welcome, of course. No, helium-filled drives won't float away if they aren't tied down.

Topic: Storage

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Vacuum? I think not.

    You would find it really hard to fly the drive heads in a vacuum and contact heads would self-destruct themselves and the platters in seconds.
    • "Vacuum? I think not."? Apparently, you flunked physics....

      Your comment shows you wrote without thinking. The drive heads would actually move better in a vacuum, much faster, with no air resistance. At the same time, the disc platter(s) themselves would be able to achieve much higher speeds, up to the theoretical rpm limit of the direct drive motors the spindles sit on. The issue of drive and contact heads self-destructing would be on the same order as they are now, more based on the heads contacting the platter at supra speeds (yes, the word is supra) in relation to standard 10K or 15K drives. However, with the alloys and composites available, a platter that can withstand contact at those speeds, as well as platters made from metal/ceramic alloys can result in a drive that will literally fly in a vacuum. The only issues after that is maintaining the drive motors (self-contained motors would work), and feeding power to the motors. This is simple, for, after all, things such as this have been done for extra-atmospheric work for years. And, for R Harris below, vacuum drives can be done, have been done, and are no big deal.
      • Link?

        "for R Harris below, vacuum drives can be done, have been done, and are no big deal."

        All conventional drives require air to float the heads. All conventional drives have a maximum altitude they can be operated at because of this. Maybe some drives have been made that operate in a vacuum, who knows. Never heard of it. Either the head positioning mechanism would be exquisite or it would be in contact with the platter which just seems impossible.

        Here is my link containing source information:


        How about yours?
      • Vacuum drives have been done?

        Like to see a reference to one. The idea has kicked around since the 50s, but I've never heard of a production drive.

        Unless you count the vacuum columns on tape drives from long ago. . . .
        R Harris
  • Good point!

    A vacuum would require a whole new head suspension system. Doubt we'll ever see vacuum drives.
    R Harris
    • SSD will take over by that point

      At least for applications where you would want a faster spinning disk.
      • On the conser market, yes.

        The transference is already well underway to SSD for OS and application storage, particularly in the mobile device market, where durability is exceptionally important.

        However, I think you're looking at a major technology jump for SSD in the server / database use. You are quite correct in that it bennifits in speed, durability, energy efficiency, heat production, and storage space. However the cost difference is huge - just on the consumer market, a 2tb performance drive is currently comparable in price to a 256gb SSD; when you're dealing in the levels of data we're talking about it's just not an option, especially when you consider many of the key advantages of ssd are actually diminished in many of these purposes.

        These helium drives are purpose built for these purposes; simple things like 4 degrees less heat production is going to make a masive difference in your cooling difficulties when you multiply the number of drives in an array, as well as reliability of surrounding components and devices.
        • 14.5 to 1 price difference

          The most economical 2TB drive I found on Newegg is $110 delivered. The same capacity in SSD was $400 per 512GB, or $1,600 total, or 14.5 times more expensive (not including space and controller ports).

          Given HAMR technology, a number of forecasts say that disk storage density improvements will outpace flash improvements for every year for the next 4-6 years, so the ratio is only going to get worse, not better.
          terry flores
  • And here I thought there was a global helium shortage...

    I think this will remain a niche market for servers that have an absolute need for speed and drive space. If this becomes the norm I don't think the helium supplies will be able to keep up.
    Michael Kelly
    • Nooo!

      I never want to live in a world without floating mylar birthday balloons... :-(
    • Helium is almost always found with natural gas deposits

      take a guess as to why helium production is down.
      • Not enough collection stations

        in Congress?
    • glut, not shortage

      There is currently a glut of helium, not a shortage.

      Regarding the 2TB for $110 at NewEgg, I bought a 3TB Seagate drive from them for $99 not too long ago. It was a time-limited special and I snagged it to put in my external case to use for backups.
  • Seems like a step sideways

    Why don't we just put the same effort into solid state drives and eliminate all the mechanical shortcomings?
    • re Seems like a step sideways

      It's the same as alternative fuels--great idea, but the petro-fuel industry has such a head start that coming up with truly competitive fuels will take decades more.

      We can't expect everyone to stop hard disk development and instead put all their resources into SSD development. As long as they keep improving hard disks, SSD's will continue to lag in capacity and cost for quite awhile, probably at least two decades.
    • Not enough flash production to replace disks

      That's why we'll continue to have disks - even if they cost the same.

      R Harris
      • really?

        here is your once in a lifetime opportunity - mortgage your house and invest in the SSD production if you truly believe that there is a shortage of the manufacturing facilities for SSD
    • One word - Profit

      As long as there is a market for very high capacity, low cost storage mechanical drives will be with us.
  • Instead of stiffer enclosure....

    ....just use a flexible membrane to equalize the volume (limited to normal sea level variations).
    Yes, to operate in an unpressurized airplane or extreme altitudes this would not work, but for 90+% of static installations this is not an issue.
  • I thought there was a shortage of helium?!?

    Other than that, think how much lighter your laptop will be!!! :-P