Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
Data Storage Power Hardware

Western Digital Touts New 'Green' Drives 119

Posted by CowboyNeal from the save-the-platters dept.
An anonymous reader writes "Western Digital today announced the availability of a new line of serial ATA drives that are supposed to use 4 to 5 watts less than other competitive drives from Hitachi GST, Fujitsu and Seagate. The new "GreenPower" line comes in 500GB, 750GB and 1TB capacities. Western Digital says it achieves better power performance by balancing the platter's spin speed in order to make it more efficient, by optimizing seek speeds and by parking the read heads when the disk is idle, according to a Computerworld story."
This discussion has been archived. No new comments can be posted.

Western Digital Touts New 'Green' Drives More

Western Digital Touts New 'Green' Drives

Comments Filter:

  • No rotational speed spec. (Score:5, Interesting)

    by niceone ( 992278 ) * on Friday November 23, 2007 @07:50AM (#21453055) Journal
    Interesting - WD don't tell you the rotational speed [wdc.com]! Must be the first drive that doesn't. In the rotational speed row it just says "IntelliPower" and below "A fine-tuned balance of spin speed, transfer rate, and caching algorithms designed to deliver both significant power savings and solid performance."
    I guess I'd need to see some independent benchmarking before I would believe that performance is not hurt. Also is the power saving dependent on the drive not being used flat out?

    • Re:No rotational speed spec. (Score:4, Interesting)

      by TheRaven64 ( 641858 ) on Friday November 23, 2007 @07:58AM (#21453087) Journal
      It sounds like the drive alters its rotational speed based on the usage. This would be really nice in a laptop drive. For example, when watching a video or listening to music, it would take less power to spin the drive slowly than to spin it up to full speed and then spin it down. I'm not sure about datacenter use. It might be that it could spin slower during periods of low demand and at full speed at other times.

      Hard drive power management is hard to get right, since spinning the drive up uses a lot of power, but keeping it spinning fast also uses some. If you spin the drive down, and then use it again, you use more power than if you leave it spinning. If you leave it spinning and then don't use it then you're similarly wasting power. Being able to spin the drive up a little bit might be a nice compromise. So would adding a large non-volatile cache.

      • Re: (Score:3, Interesting)

        by moosesocks ( 264553 )
        For that sort of scenario, you'd use the least amount of power if you cached the entire movie (or as much as you could fit) into memory in one fell swoop. Spin up, read the entire file, and spin down.

        I seem to recall that one of the ways in which Apple tweaked the battery life of the iPod was to considerably increase the size of the RAM cache, and read as much of the playlist as possible into memory.

        • Re: (Score:3, Insightful)

          by TheRaven64 ( 641858 )
          That would, of course, give you better battery life. Even better would be to cache the entire contents of the disk and never spin it up...

          In the real world, even if you have enough RAM to cache the movie, other things also want to use some of it. The demands for streaming it from the disk are quite low; around 1-2MB/s, while my laptop's disk can do between 10-30MB/s in sustained transfers. Dropping the speed from 5400RPM to 540RPM would allow this demand to be met without swapping anything out and causi

        • Re: (Score:1)

          by zergl ( 841491 )

          For that sort of scenario, you'd use the least amount of power if you cached the entire movie (or as much as you could fit) into memory in one fell swoop. Spin up, read the entire file, and spin down.

          I seem to recall that one of the ways in which Apple tweaked the battery life of the iPod was to considerably increase the size of the RAM cache, and read as much of the playlist as possible into memory.

          RAM needs quite some power to operate as well (at least as long as we're stuck with DRAM, with all that a

        • Unfortunately the hard drive has no way of knowing how much data a program is actually going to read from a particular file, so this is not something the hard drive firmware can handle.

          It would be an issue for the software developers to determine the benefits of caching a huge file into RAM, and from their point of view there is little point in doing that for data you're just going to read linearly.
          =Smidge=
          • an OS is quite capable of performing read-ahead.
            • Again, that doesn't help the hard drive manufacturer, and the OS does not know how much data other applications may want to read at any given time. Only the application author can know that... maybe.

              =Smidge=
              • That's not how we write OSes. if you read from a file we will read ahead on that file.
                • That's great... but by how much do you read ahead?

                  How can the OS know if the amount to buffer is optimal for that particular application?
                  =Smidge=
                  • application can control the per file buffering recommendation. or they can use the default. what do you think happens when you use setvbuf() is altered? or when a program is smart and just mmaps the portion of the file they are interested in?

                    • application can control the per file buffering recommendation.

                      Bing!

                      And if you don't realize this is exactly what I said in the first place (ie only the developer of the specific application knows the optimal buffer size), you need to be held back a grade. The OS can not magically know this, it needs to be told. The hard drive, also, can not know this, and there is currently no mechanism to tell it. The point being that if you COULD tell the hard drive what kind of demand to expect in the long term, it could

      • An interesting idea, but they better implement it correctly: keep reading while changing the speed. With high-speed CD drives I always get the impression they only start reading after the drive has arrived at top speed, which means that for small files a 40x drive ends up being slower than a 4x drive.

        • Re: (Score:2)

          by tknd ( 979052 )

          Hitachi has already implemented the "incorrect" version for a number of years now:

          From their Feature Tool User's Guide PDF [hitachigst.com]:

          APM is a technology that adaptively adjusts the power saving feature of New Hitachi and Legacy IBM hard disk drives to suit your working style. Power is saved by idling the actuator arm, additional power is saved by unloading the actuator arm and heads from the disk to the off-ramp. The technology actively adjusts the trade-off between disk delay and power consumption. This program

          • I don't know enough about hard drives to know why they don't have variable read speeds and why the head needs to be moved off of the platter when the disk spins down

            The heads have to be moved off the platter(s) during spindown so they don't crash. The heads "fly" breathtakingly close to the platters only when the platters are spinning at speed.

            I imagine variable speeds with heads loaded isn't done for the same reason: Probably a fairly tight range of RPM that will keep the heads flying properly.

      • Re:No rotational speed spec. (Score:4, Insightful)

        by PopeRatzo ( 965947 ) * on Friday November 23, 2007 @10:38AM (#21454111) Journal
        Can't we just go straight to the solid-state memory and stop worrying about "spinning disks"? I mean, think about it: very soon, we're going to laugh about the fact that we used to use these boxes with spinning platters inside to save all of our data.

        Personally, I'm not going to be making any further large investment in any storage media that has moving parts. I'll replace drives as they die in my little RAID box, but that's it for me.

        I look down at the little 8gig Sansa mp3 player hanging around my neck when I ride my bike and I think: this little thing is pumping wattage into my cottage, rocking my head at serious volume and it runs for 20 hours on a one-hour charge, and I can fit the Herbert von Karajan recording of Wagner's Parsifal like 20 times over and still have room for a few movies, and there's scars all over the case from having bounced it off the pavement countless times, but it works like a charm. This has to be the future.

        • Re: (Score:2)

          by d'fim ( 132296 )
          I can't recall if it was one of Peter Norton's books, or in the preface to PC Magazine's _DOS_Power_Tools_, but back in the heyday of DOS 3.3 the observation was made that an HDD is an aluminum platter coated with iron oxide; and the author then asked "would you trust your data to a rusty pie plate?"

        • Re: (Score:2)

          by Yvan256 ( 722131 )
          Unless there's some incredible breakthrough coming soon, at prices that can at least match the old spinning disks technology, we're not about to replace TeraByte drives which are needed to store DVD and CD rips.

          So yes, solid-state storage is the future, but not until it can match the capacity/price for massive amounts of storage. It's got the low-capacity (i.e. let's say 8GB for now) covered, but the high-capacity (let's say over 40GB) is still the domain of spinning disks.

        • Re: (Score:1, Insightful)

          by Anonymous Coward
          There's a post like yours in every /. thread about flash or hard drives.

          Yes, having flash for bulk storage would be nice. It'd also be nice to have main memory run at the same clock speed as our CPU instead of having this complicated caching system. But we do caching and use spinning discs because they're cheaper, and as long as both flash and spinning discs improve at the same rate (roughing Moore's Law) and our demand for space keeps increasing, we're going to keep on using spinning discs for bulk storage

          • in ten years, the 512 GB flash drives will be $100, but the 20 TB hard drive will also be $100, so we'll still have both.
            don't be too sure. My basement is littered with technology that we'd thought would be around forever.

      • Re: (Score:2)

        by mmontour ( 2208 )

        It sounds like the drive alters its rotational speed based on the usage.
        That's exactly what WD's marketing department wants you to think, and the previous text on their website implied that even more strongly than what's currently there. They now concede that the rotational speed only varies *between models* in the "Green Power" line, and multiple tests (including my own) have shown that the rotational speed of the 1TB model is at or very near to 5400 rpm.

      • Re: (Score:2)

        by pdbaby ( 609052 )
        I own one of these disks and I got the impression from WD that they run each disk at a constant speed based on the optimal speed for that individual drive because the motor would be more complicated to run at a variable rate - although I'm not a mechanical engineer, so bury me in salt before listening to that :-). What I do know is that it runs 10C cooler and significantly quieter than the other drives (Seagate) in my Mac Pro. The motors on the WD GP are definitely a lot quieter, especially when spinning u
    • According to some (probably marketing) sources, the rotational speed varies between 5400 and 7200 RPM. I've only seen this 2nd handed by now, trying to locate the original source; will get back if I do.

      • Re:No rotational speed spec. (Score:5, Informative)

        by darthflo ( 1095225 ) on Friday November 23, 2007 @08:27AM (#21453225)
        I can't seem to find anything coming from WD mentioning RPM, so the data from some online stores near ("IntelliPower at 5400-7200 RPM) me isn't quite verifiable. According to independent reviews, the drives seem to clock in somewhere between those values, so those might be the theoretical upper and lower limits, respectively.

        Assuming the [Green Power] also shares such a seek time, that leaves us with 15 ms [measured access time] minus 9.5 ms [assumed seek time] which equals 5.5 ms, almost exactly the rotational latency associated with a 5400 RPM spindle speed.
        (from storagereview.com [storagereview.com])

        [I]t's easy to convert [WD's values for average rotational latency] to revolutions per minute, or RPM. 5.6 milliseconds of rotational latency works out to about 5,400 RPM, which just happens to be the low end of the GreenPower's spindle speed range. Western Digital says that's by design; the latency spec it lists in the GreenPower's data sheets is merely an estimate based on the spindle speed range of the drive.
        (from techreport.com [techreport.com])

        Aside from those missing values, the drive's power consumption (4W idle, 7.5W read/write) seem pretty nice compared to the rest of the market.

        • Re: (Score:1, Offtopic)

          by Pulse_Instance ( 698417 )
          Drop down lists are hard before coffee, is there any other way to undo a bad moderation with the new system?

          • Re: (Score:3, Informative)

            by ColdWetDog ( 752185 )

            Drop down lists are hard before coffee, is there any other way to undo a bad moderation with the new system?

            Don't moderate before coffee. In fact, don't do ANYTHING before coffee.

        • Re: (Score:3, Insightful)

          by tomhudson ( 43916 )

          TFA says it *might* say you "up to $10/year" on electricity.

          In other words, YMMV, etc.

          So lets look at REAL figures.

          I'll save 4 watts per drive on my 4-drive home box. 16*24*365/1000 = 140 kw/hours.

          @ 7 cents/kwh, I'll save a grand total $9.80 for 4 drives, or less than $2.50/drive.

          That's if I run it 24 hours/day.

          Most likely consumer-use scenario is less than a buck a year. I'll leave my drives spun up at full speed all the time, thank you. Easier to save a LOT more money just by turning off the l

          • If you're just looking at energy/money savings, it will, of course, seem pretty pointless. Before WD's "Green" drives hard drives didn't use up much, after they don't use much less (because they didn't use up much in the first place).
            On the other hand, if those drives actually save an average of 5W when compared to the competition, I find that a pretty damn impressive feat. Put into perspective (4W idle, 7.5W seek; assuming about 30% of total running time spent is seek/read/write), that's like Ford's newes

            • What about heat? (Score:3, Interesting)

              by Sensible Clod ( 771142 )
              Another important benefit for such an improvement would be for cool/quiet computing, which is high on my list of desired system attributes. To put it simply, it's like taking a 4 or 5 watt heater out of the machine (which should also make it last longer).

              I'm a little concerned about parking those heads all the time, however. Last thing I need is a cool-running drive with worn-out ramps...

    • Re: (Score:1, Informative)

      by Anonymous Coward
      I tried one of these. Performance on large files is about the same as the seagate barracuda 750GB. Performance on small files, or when there's a lot of seeking to be done, or under heavy load (which requires seeking) is noticeably slower. Maybe 75% of the barracuda, as a off-the-cuff completely unscientific number.

      The power is noticeably lower than the barracudas, even comparing an active one of these new WD ones versus an idle barracuda.

      Overall I'm very happy with it. Those barracudas cooked - definite

    • Re:No rotational speed spec. (Score:5, Informative)

      by MrNemesis ( 587188 ) on Friday November 23, 2007 @08:14AM (#21453169) Homepage Journal
      From what I remember from the reviews, the disc spins between 5400 and 7200rpm depending on load. Benchmarks showed it's not as fast as equivalent 7200rpm drives of the same capacity, but the performance disparity is in the region on 5-10% at worst. For people worried about power usage and/or noise though, it looks like a superb drive - perfect for an HTPC.

      http://techreport.com/articles.x/13379 [techreport.com]

      • Re: (Score:2)

        by Fweeky ( 41046 )
        Variable RPM during usage would be tricky; head flight height would be all over the place. Everything I've seen suggests it's fixed somewhere around 5400RPM, but is capable of throttling down when not in use so when it wakes up it doesn't have to spin up from 0RPM.

        I wonder why WD don't seem to publish a read bit error rate for these drives. Seagate's 7200.11's rate 1 sector per 1^14, or 1^15 for the ES.2; neither of which are particularly wonderful (what's that, 1 error per 10-100TB read?), but you'd hope
        • IIRC WD have brought out an "RE2" edition targeted at the enterprise market, as they do with mose of their range. Just been poking around WD's site and, as you point out, couldn't find a spec sheet listing the read error rate for any of the drives I looked at, which is a bit annoying.

          Pretty sure there is a variable spindle speed though; is it possible they alter the height of the head during travel, depending on spindle speed...? Sounds quite tricky but surely some clever electricla engineering could solve

        • Re: (Score:2)

          by tlhIngan ( 30335 )

          I wonder why WD don't seem to publish a read bit error rate for these drives. Seagate's 7200.11's rate 1 sector per 1^14, or 1^15 for the ES.2; neither of which are particularly wonderful (what's that, 1 error per 10-100TB read?), but you'd hope WD are closer to the latter than the former if they're touting these for enterprisey usage.

          I've always had a question about that... is that reading the data off the media (and thus corrected before it hits the external bus), or is that an undetected error that makes

    • This reeks of the 4500 rpm Quantum Fireballs of old. They can pull off the bit density but not the corresponding channel throughput so they have to slow down the rpm to avoid wasting capacity. The Green thing is marketing's way of selling it.
    • I first learnt about WD's green drives from their site and I remember they clearly explained the RPM rate spectrum of a green drive. I check now the link you provided and I don't see the rate. Did they remove it? Why? Whose marketer's idea was this? As a customer I would never consider a dynamic RPM drive if I don't know the rate range, no matter what they call it. Even though the rotational speed may change, it still has got to have a range, eg between 5400-7200 etc, and as consumers we want to know
    • You can tell based on the seek time of 8.9ms, that it's around 7200rpm mean.

  • How Green is Green? (Score:3, Insightful)

    by Anonymous Coward on Friday November 23, 2007 @07:55AM (#21453071)
    It's good to see manufacturers trying to reduce power consumption in their products, and I hope the trend continues (without impacting performance). However the big savings are more likely to be found in the manufacturing processes. How much energy could be saved there? How much "greener" could the chemical processes be?

    It's neat, it's a start, I'm sure it'll produce a decent amount of ad copy for them, but it's not really very "Green".
  • I can see that developers are getting round to the idea, but every engineer should always also optimize for energy effeciency. For the Fritzbox router that I have http://www.avm.de/de/Service/Service-Portale/Service-Portal/index.php?portal=FRITZ!Box_Fon_WLAN_7170 [www.avm.de] you can get a beta firmware that has savings features enabled, I think that's a great idea.
  • now where is my paint.

  • Solid State? (Score:4, Interesting)

    by Ginger Unicorn ( 952287 ) on Friday November 23, 2007 @08:23AM (#21453205)
    How does this compare to solid state drives in terms of power efficiency?

    • Re:Solid State? (Score:4, Insightful)

      by Kjella ( 173770 ) on Friday November 23, 2007 @09:24AM (#21453581) Homepage
      Still not close, but I think they're competing in different markets. If you need 500GB disks, you're not looking at SSDs (unless you got a spare ten grand or so).

      • Re: (Score:2)

        by gmack ( 197796 )
        At the price flash drives are getting to it's tempting to get a SATA to CompactFlash adaptor and put the base system (where performance matters most) on that and keep the 500gb drive for logs and movies..

        So they may not be entirely different markets.

        • Re: (Score:2)

          by tlhIngan ( 30335 )

          Can't we just go straight to the solid-state memory and stop worrying about "spinning disks"? I mean, think about it: very soon, we're going to laugh about the fact that we used to use these boxes with spinning platters inside to save all of our data.

          Problem is, storage capacity has risen faster tha Moore's Law, thus, spinny disks get more storage faster than solid state memory gets more memory. Also, people want large disks. A 64GB SSD costs the better part of a grand, and people want it in their laptops t

        • At the price flash drives are getting to it's tempting to get a SATA to CompactFlash adaptor and put the base system (where performance matters most) on that and keep the 500gb drive for logs and movies..

          What about the "number of writes" property of magnetic disks. I can write to a drive hundreds of thousands nay millions of times whereas with a Flash you might get ten thousand if your very lucky. I've head of flash drives being used as boot disks for small media centres and they die after a month simply

          • Re: (Score:2)

            by gmack ( 197796 )

            What about the "number of writes" property of magnetic disks. I can write to a drive hundreds of thousands nay millions of times whereas with a Flash you might get ten thousand if your very lucky. I've head of flash drives being used as boot disks for small media centres and they die after a month simply because the o/s writes to the drive very often.

            Well that's why I put the logs on the 500gb drive.

            Funny thing though.. my dad used to have Toshiba T2000 with a 4mb flash that the laptop used as ram(yes

    • You want to compare a 500GB hard drive that uses 5-15W against a SSD of 32 or 64GB that uses only 2 to 5W? You're welcome
      • thanks for your approval, now i feel validated as a person. Perhaps I only care about power usage rather than capacity. Ever consider that?
        • Please excuse my harshness.
          You will find here a comparison of HDD power use (and others characteristics):
          http://www.anan/ [www.anan] dtech.com/printarticle.aspx?i=2982

          Power Draw Idle / Load .16W / .48W .87W / 2.42W 9.19W / 10.02W
          SSD 16GB 2.5" Laptop Raptor

          So, a 5W reduction in power would bring a 500GB 3.5" HDD po
        • Power efficiency?
          compared to a 160GB laptop drive, you end up with:
          about a fifth of power use
          a tenth of capacity
          probably a lower weight
          no noise
          half the average speed and a fifth of burst speed
          depending on scenario, somewhat lower performance to a fifth of the performance
          As for power efficiency, you have:
          Half the capacity/power use
          Better performance/power (from somewhat lower to more than 5 times better, depending on the scenario)
          More $ multiplied by Watts
    • A word to the wise, stay away from the Solid State society, or you'll end up killing yourself just like them.
  • Marketing people in technological companies often have no knowledge of technology and don't think it is important to have any.

    Western Digital GreenPower Hard Drives [westerndigital.com] seem to be heavily influenced by that attitude. There seems to be no information about actual speed, giving the impression that arrogant marketing people have decided that technically knowledgeable people can be manipulated, and won't notice that lack of specifications.

    It's difficult to compete with Seagate's 5-Year Warranty [seagate.com], reliability i

    • Re: (Score:2)

      by NMerriam ( 15122 )
      Let me guess, you finally get up the nerve to ask out that cute girl in marketing and got turned down?
    • It's difficult to compete with Seagate's 5-Year Warranty If the 1TB WD drives are actually reviewed and available, though, that would be a big advantage over Seagate and Samsung's vaporware.
  • Look, WD, I don't care what color you spray paint your hard drives. After my failed HD was replaced by another that started clicking almost immediately, I ain't never going back.
  • Yes. I'm sure WD is green. Their failure rate makes them some of the least power-consuming drives in the industry. Well done for being 10 years ahead of the curve.
  • Something I have not seen discussed is that 40% less power means 40% less heat generated. Potentially, this means fewer fans (and less fan noise), lower air conditioning costs if you have a lot of drives in a data center, and longer electronics life for the drive.
  • I really hate the term "green" when used to describe a products so-called friendliness with the environment. Just about every computer product touted as green has some form of toxic substance contained within it or uses nonrenewable energy to function. Anyhow, kudos on the HD power reduction technology by WD, but don't call it "GreenPower".
  • Is in Green in the same way that a 500GB disk holds 5,000,000,000 bytes (== 465 GBytes) rather than 536,870,912,000 bytes ?
  • can I get one in blue?
  • So you'll see the savings in about 5 years... Hold onto the money or buy a carbon credit. This is a feel good purchase in the first round, the best part about buying it now would be voting with your dollars that this is a technology you want to see more developed.

Slashdot Top Deals

A committee takes root and grows, it flowers, wilts and dies, scattering the seed from which other committees will bloom. -- Parkinson

Close