back to article Seagate promises to HAMR us all with spinning rust next year

Seagate has revealed fresh data on how it will increase areal density as well as a disk's IO rate from next year onwards. Analyst haus Stifel Nicolaus' MD, Aaron Rakers, attended an analysts' day at which Seagate said it is learning about potential caching techniques to ameliorate the write performance disadvantage of shingled …

  1. Your alien overlord - fear me

    "Seagate said it is learning about potential caching techniques to ameliorate the write performance disadvantage of shingled magnetic recording (SMR) disk drives." - WTF are they on about?

    1. Bronek Kozicki

      in other words, they are researching some kind of presumably write cache ("learning about caching techniques") which could be used to skip or shorten "rewrite all the data" step, that is necessary when storing the data in shingled hard drives. Basically a message to market analysts (those who are familiar with SMR).

    2. Gordan

      (Most) SMR drives have many zones, a staging zone. All the writes first go sequentially into a staging zone (append-only, similar to a log). When the drive is otherwise idle, this data gets committed into the target zone (which requires rewriting the entire zone).

      So if the load profile on a SMR drive is read-mostly (and the vast majority of typical load profiles are in fact read-mostly), the performance isn't too bad most of the time. In fact, with the staging, random writes get merged into one big sequential write, which avoids some of the write-time seek latency.

      However, if the staging area is full and the disk hasn't had enough idle time to commit the data from the staging zone(s) to the target zones, it will have to do so before it can accept more writes, at which point the write performance drops to less than 1/3 (as of spinning rust wasn't slow enough already). Having said that, SMR drives seem to have pretty large staging areas so this doesn't happen particulary often in typical use.

  2. PleebSmash
    Thumb Up

    acronym excess

    At least HAMR is a funcronym.

  3. Anonymous Coward
    Anonymous Coward

    SMR from my own experience means seagate get to make a larger capacity drive, sell it with a 1 year warwanty. But due to all the frequent and unnessicary rewrites because of re writing a shingle block it doest last as long. Where as non smr drives would re-write just the bit that changed. Its a great idea for increasing capacity, but from what ive observed at the cost of drive endurance (life expecancy) when used hevily, not to mention speed.

    1. Gordan

      @AC:

      There is in fact no correlation between how heavily a mechanical drive is utilized and the probability of it's failure. You can find Google's disk reliability study on this subject here:

      http://static.googleusercontent.com/media/research.google.com/en/archive/disk_failures.pdf

      (Section 3.3)

      1. Anonymous Coward
        Anonymous Coward

        Your 2007 report would not likely see the full picture the dm (shingled) range were fairly new to the market then if out. Post that report i saw most faliures at less than 2 years of age with dm's. I belive the reason for this was maybe the ext4 file system with journalling... Its an issue for a very high number of drives i saw running ext4. Drive issues, were rare for me pre dm series, then any thing that had heavy disk read \ writes use and used ext4 started to fall like rain, switched to wd and had 0 bad drives in 3 years. Intrestingly you will see at the inception of the dm series seagate started to massively reduced the warwanty periods for drives, makes you wonder. And pre the report drives did last fairly well, most seagate faliures i saw were shingled drives or for the pre shingled drives a few with bsy issue:

        https://sites.google.com/site/seagatefix/

        1. Gordan

          @AC:

          Actually, the discrepancy in reliability between different manufacturers and drive generations easily more than covers your perceived failure pattern, and there is still no evidence that it has anything to do with the drives working harder, it is almost certainly entirely down to the disks from some manufacturers being massively more crap than disks from others. Since you mention "DM" drives, I presume you are talking about Seagates, whose most reliable model has several times the AFR than, for example, HGST's least reliable model. See here for the Backblaze study on this:

          https://www.backblaze.com/blog/what-hard-drive-should-i-buy/

          Based on the large number of disks in the sample and relatively extensive analysis, coupled with prior evidence that the workload of the drive has no effect on longevity, strongly implies that a drive being SMR doesn't really impact it's reliability, over and above by the fact that it is a less mature technology (bleeding edge drives are usually less reliable than the longer term established product lines that have had manufacturing processes perfected and bugs ironed out over some time).

      2. Joe Fagan

        @Gordon

        Section 3.3 shows a very strong correlation between duty cycle and AFR!!!

        Thanks

        Joe

        1. Gordan

          @Joe:

          From the paper:

          "... only very young and very old age groups appear to show the expected behavior. After the first year, the AFR of high utilization drives is at most moderately higher than that of low utilization drives. The three-year group in fact appears to have the opposite of the expected behavior, with low utilization drives having slightly higher failure rates than high utilization ones."

          So only very young drives fail faster when heavily utilized, implying that if they were already marginal when they left the factory, hammering them harder will finish them off quicker. Among the 1-4 year old drives the difference is minimal, and in case of 3 year old drives, it is the low utilization ones that are more likely to fail.

          So the correlation is tenuous at best for the majority of drive ages. Also note that among the very young disks, the figures show that low utilization disks were also 2-3x more likely to fail than medium utilization disks.

          So overall, not that much of a correlation.

          1. Joe Fagan

            Gordan,

            The sentence and the graph just don't 'correlate'! The AFR for high utilization is on average about 2.5x versus low. You'd have to take yr 3 data as an anomaly - the text talks about survival and different models as possible explanations. Thanks for the link - very interesting.

            Thanks

            Joe

            1. Gordan

              @Joe,

              Well, differences in the 1, 2, 3 and 4 year old disks are all within 1% of each other, so there is no strong signal there either way, i.e. it makes negligible difference. Compared to that, < 1 year and 5+ year entries could be viewed as anomalous. I guess it comes down to personal interpretation, but it certainly comes across as not being something worth worrying about in the 1-4 year age group.

  4. Gordan

    Two Heads Per Platter

    This has been tried 20 years ago by Conner, in their Chinook line of disks. It was uneconomical compared to the alternatives back then, and I don't see that it'll be any different this time around.

    Spinning rust is nearly dead, anyone saying otherwise is only doing so because they are selling it.

    1. phil dude
      Thumb Up

      Re: Two Heads Per Platter

      I'm inclined to agree with this sentiment, especially when confronted by the rapid advances (and the extra dimension!!) in flash...

      The real issue is, what place in the market will disks occupy? Obsolescence is almost mandatory in technology, but it is difficult to imagine a place for spinning rust when you can have flash, *other* than price...

      Until very recently, you couldn't *buy* the flash equivalent (due to size limitations) but now we are seeing direct competition emerge...

      P.

      1. PleebSmash

        Re: Two Heads Per Platter

        Back when NAND was running to an endurance wall, it seemed like HDDs could linger on. Now 3 bits-per-cell vertical NAND is mainstream and you can get a bigger 2.5" SSD (15 terabytes) than 3.5" HDD (10 terabytes).

        WD and Seagate have to bring the HAMR down and deliver drives at $10-20 per terabyte and lower. They also have to ship NAND themselves. Hybrid drives can still be a good idea, especially if every hard drive they sell ships with NAND.

        1. Gordan

          Re: Two Heads Per Platter

          @PleebSmash

          "Back when NAND was running to an endurance wall"

          That was never actually the case on SATA SSDs, it was merely perceived as such when people reacted with a kneejerk instead of looking at actual data and usage patterns. The endurance of NAND has actually been reducing steadily with the process sizes, it has not been improving, despite ongoing incremental advances in flash management. What has happened is that people have begun to understand that quoted write endurance is ample for any sane use over the most extreme foreseeable useful life of the drive (e.g. 10 years) even if you can only write 75TB of data to a 1TB SSD. Persisting 75TB of data in anything resembling desktop, or even typical server use, is way out there in the statistically insignificant territory fraction of use cases.

          Then consider that even under very harsh tests every SSD tested to destruction has managed to outlive it's manufacturer specified write endurance by a large multiple:

          http://techreport.com/review/27909/the-ssd-endurance-experiment-theyre-all-dead

          Unless your use case is large scale very ephemeral data caching (high churn caching of TBs of hot data out of PBs of total data) or runnaway logging, SSD write endurance is way beyond being worth even remotely worrying about. It is, and always has been, a non-issue for most sane use cases.

    2. Joe Fagan

      Re: Two Heads Per Platter

      Gordan,

      TDMR is not about two actuators. It's about 2 or more read heads on each slider to increase SNR. With multiple read heads looking at almost the same position, random noise can be averaged out (if they're positioned in a line one behind the other) and intertrack interference can be minimised (if they're positioned parallel one beside the other) - so track density can be increased for the same net SNR.

      There is no possibility of reading different random tracks as the article suggests.

      As for the demise of HDD...

      In the performance/reliability/cost equation, cost is king for the bulk of data stored in the world. The very fast, some very reliable and the very small is transitioning to flash. While people continue to care about cost, the bulky will continue on HDD. The bulky is the category that continues to grow much faster the the others - it's cloud, IOT and object storage.

      Thanks

      Joe

      1. Gordan

        Re: Two Heads Per Platter

        Yes, spinning rust is still cheaper, but the gap is ever narrowing. But cost also increasingly includes power consumption, but for long term storage of enormous amounts of write-once read-never (or near enough) data, NAND is increasingly becoming a contender. One example of a point in case being here, and that was more than 2 years ago, certainly long enough ago that it may well be widely deployed today:

        http://www.theregister.co.uk/2013/08/13/facebook_calls_for_worst_flas_possible/

        Which leaves spinning rust being consigned to cheap desktop grade systems as the only prospect it has to look forward to.

        I do still use spinning rust in some bulk storage applications, but I rather expect to be replacing them with SSDs (as I have been doing of late) as and when they expire out of warranty. Then again, I only use HGST drives so the expiry will likely take years rather than months.

        1. Mellon

          Re: Two Heads Per Platter

          About 4% of the bits shipped last year were flash. Total bits shipped grows by 20+% per year.

          Think of the investment needed to fab those 4%. Flash will be doing well to maintain that percentage given the capital expenditures required to increase wafer output.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like