Comments Locked

27 Comments

Back to Article

  • Snoopykins - Monday, September 9, 2013 - link

    Very nice article, I always love the ones regarding new ideas to push beyond current limits. I for one would be interested in seeing some more real world numbers. If Anandtech could find what drives are currently using SMR and compare them price and performance wise to the ones that aren't, that would make for a pretty helpful article. It might help people decide to buy them or not, and it might also help put a finger on what the pricing is going to be like on the larger capacity drives of the future. Your final sentence was "The real question is whether or not Seagate can maintain similar full drive performance compared to a non-SMR drive." I think you, Anand, of all people have what it takes to find out!
  • melgross - Monday, September 9, 2013 - link

    The question is whether an SMR drive can be differentiated from a "regular" drive from testing. If the rewriting process can be detected, then maybe yes, but if not, who knows?
  • Snoopykins - Monday, September 9, 2013 - link

    He could find out what ones are SMR from Seagate or by other means, then compare the prices. For performance, if he can't find a way to find a difference in performance, then we probably won't either, and therefore they are equal in real world usage scenarios.
  • xdrol - Monday, September 9, 2013 - link

    This process is not much different from the one SSDs use, you can detect it there, why could one not detect it here?
  • name99 - Monday, September 9, 2013 - link

    Depends on how well Seagate works to hide it.
    For example if they add 4GB of invisible flash, store immediate writes to the flash, AND can simultaneously write to flash (incoming stores) and read from flash while writing to disk, then it could all work out just fine.

    It's not impossible to make this transparent, it's just a question of whether they did or did not do the job well.
  • Kyle_PL - Friday, December 13, 2013 - link

    This "invisible flash" would be killed very fast ... and the price of this HDD would be higher, so there is no sense for your solution.
  • KAlmquist - Monday, September 23, 2013 - link

    The effect of SMR should show up in random read vs. random write performance. Consider what happens on a SMR drive when you write a block in the fourth track of a band. You have to read the three tracks which come before the track you are interested in, then write the block you actually want to change, and then rewrite (portions of ) the three earlier tracks. That's seven physical I/O operations, so the disk has to rotate six times after the first I/O operation, which takes 50ms on a 7200 RPM disk. The bottom line is that random writes are going to be much slower than random reads.

    The above analysis only applies after the benchmark has reached steady state. To reach steady state, you have to write enough data to force the drive to actually write the data to the disk rather than storing it in cache. If the disk uses RAM for caching, that's not a big deal, but if the drive has 8GB of flash memory, it will take a long time to fill up the cache.
  • Cow86 - Monday, September 9, 2013 - link

    The idea is valid, but I can imagine this will reduce write speeds on an already filled disk-section....because it'll have to do the entire band, instead of just overwriting the requested data. So slower write speeds in the later (filled) life of a HDD should be expected I would think?
  • Azethoth - Tuesday, September 10, 2013 - link

    No, on average it has to rewrite half the band. It is only worst case that rewrites the entire band (change at start). Best case is no rewrite for changes at the end of the band.
  • Cow86 - Tuesday, September 10, 2013 - link

    Ah right...I sort of misunderstood...still, you'll have to rewrite more on average than just the desired new data...so slower speed on average? Would be interested in seeing tested how much of an impact it might have as well.
  • futrtrubl - Monday, September 9, 2013 - link

    I would also be interested to hear what the effects of this on drive life will be considering you now have SSD like write amplification.
  • melgross - Monday, September 9, 2013 - link

    I'm not so sure it does. Rewriting a HDD doesn't weaken any of the magnetic surface the way cells are weakened in flash NAND.
  • chaos215bar2 - Monday, September 9, 2013 - link

    That's not what write amplification means.
  • xdrol - Monday, September 9, 2013 - link

    That doesn't, but rewriting means you need to write more, and more writing means more mechanical wear, at least for the head positioning mechanics.
  • cygnus1 - Tuesday, September 10, 2013 - link

    Probably not. I would imagine the head doesn't have to move at all to read or write a single "band" with this tech. It simply has to read the band prior to rewriting it with the changed data. Also, with enough NAND write buffer on the drive it may be incredibly improbable to actually hit the performance degradation imposed by this tech.
  • madmilk - Monday, September 9, 2013 - link

    Probably minimal. Hard drives wear when the head moves around a lot, but this just makes every write into a sequential write for the whole band. The head doesn't get banged around like with random I/O.
  • glugglug - Monday, September 9, 2013 - link

    Does this mean defragging the drive will have minimal effect because there is an extra mapping between logical and physical sectors?

    Also, I would think typical Flash wear-leveling algorithms would be unsuitable because of high random access times, so the write amplification factor once the drive is filled is probably higher than an SSD. Does the drive reorganize the data to mitigate this during idle time?
  • djc208 - Monday, September 9, 2013 - link

    Though even with the issues in the article and the discussion, it should still dovetail in nicely with SSDs. This as a main drive may come with a performance write penalty, but with an SSD for programs/OS, and this for pictures, video, etc. where the random and small file writes would be minimized it would become even less of an issue.
  • alexvoda - Monday, September 9, 2013 - link

    HAMR seams like a much better idea than this.
    Both achieve higher densities but SMR sacrifices quite a lot.
    The only advantage HDDs have over SSDs is the cost per data storage unit.
    Adding some of the disadvantages SSDs have to HDDs will not do the technology any good as SSD prices are falling.
  • xdrol - Monday, September 9, 2013 - link

    HAMR and SMR are not exclusive technologies, a drive could have both.
  • piroroadkill - Monday, September 9, 2013 - link

    This seems dodgy, and the increase in density is not exactly mind blowing. Not too sure about this.
  • name99 - Monday, September 9, 2013 - link

    If you think this is "dodgy" then I'd advise you not to look at how HDDs have operated since at least 2000. Concepts like PRML and FEC will undoubtedly make you cry.
    (And you may want to also not look into how optical disks have ALWAYS operated, into how GigE operates [trellis coding to provide 6dB FEC gain], or how any wireless spec more recent than 802.11b operates.)
  • iwod - Monday, September 9, 2013 - link

    My HDD is already fast enough for everything i intend it to do. Otherwise i have a SSD. And if SATA Express and PCI-Express 3.0 SSD could come faster the speed difference would be even greater.

    I want bigger HDD. Not 5TB but something more like 6TB by 2014. We haven't had much improvement in terms of capacity since the flood. I suspect the HDD makers are trying to milk the market as much as possible given their relatively low margin of profits.
  • Urizane - Tuesday, September 10, 2013 - link

    I hope you realize that at 6TB you're asking for 48 trillion 1s and 0s. Perhaps an external enclosure with multiple drives would be more sensible, anyway. You would be better off protecting your data (RAID 5, 1+0, etc.) if you need that much of it. At that density, I wouldn't trust bits to be stable for very long (think quantum effects and so on).
  • NetMage - Tuesday, September 10, 2013 - link

    I wonder what happens if a power failure occurs at a bad time.
  • jimdibb - Wednesday, November 27, 2013 - link

    You really have to understand that these drives are not for your desktop. They are for archival, static storage. As used here http://www.theregister.co.uk/2013/11/26/facebook_d...
  • Alientech - Friday, April 4, 2014 - link

    Well they are already shipping. The write speeds are extremely slow and on par with USB2. Read speeds are the same pretty much. The Zones seem to be in a U shaped bit density where the edges have much higher density than the middle. The new ST3000DM001-1E6166 drives with the FC4x firmware have them. Not exacty sure how the writes work but they start off at 150MB with the cache empty and soon goes down to under 50MB after like 10GB or so of data. The speeds just dont make any sense and neither does the HDtune graphs with each zone showing a U shaped transfer rate. The last zone speeds drop down to around 80MB while on the older drives it was closer to 100MB. Average read speeds are down from 170MB to 150MB.

Log in

Don't have an account? Sign up now