The fundamental decision facing most of the online storage industry is whether to use spinning or flashing bits or both. Flashing bit growth, solid state NAND storage, is booming and set to grow strongly. Why aren't Seagate, Toshiba and Western Digital, the spinning disk-supplying tri-opoly, as successful with flash drives as …
I always thought this was bullshit. Either you make/sell the "new stuff" and your business evolves with the times and you actually stay in business, or you don't, and people simply buy the new stuff elsewhere.
Of course there might be the problem you don't have the skills/technology to make the new stuff because it's so different from your core business, but that's a different problem.
they might be working on the next flash iteration
Is it really worth for these companies to buy controller technology for NAND flash ? NAND seems to be at the end of its useful life anyway so in 5 years time, chances are flash drives will use some other technology which will have other physical constraints that the controller software will be asked to overcome. As the article points out, there is no rush yet. Maybe the right strategy is to bet on the next tech and develop it internally, rather than joining the fight in the current flash iteration.
"NetApp has said it will cache data from its arrays in server flash cache and will need SW to do that."
NetApp has been doing this for a few years, so maybe they have been doing it without software until now.
Five year's time ?
As the article says, flash production will take many years to build. Until then the price difference alone will ensure storage keeps spinning.
If floppy disks were cheaper than USB sticks, just as fast and larger capacity, we would still be using them.
The Writing's on the Wall for HDDs
I work for a fairly small outfit so you may have to scale accordingly, but we're already buying far more SSDs than HDDs. In fact, we haven't deployed a single desktop with an HDD since early 2009 and since then we've purchased many dozens of SSDs. In that same time period we bought exactly 7 hard disks. The HDDs we still have which predate 2009 should all be put out of service by the end of next year. This includes disks in our servers. In fact, we only use HDDs for bulk storage (and even that array has an pair of SSDs for caching). 15k rpm drives don't offer any gains over cheaper SSDs for virtualization or even server OS disks (RAID 1 is your friend).
At some point we may need to expand our storage or simply replace the aging NAS array, but that might mean at most another 7 drives or so in the next 5 years. During that same time we will undoubtedly buy many more dozens of SSDs (and maybe even a unit or two of whatever comes next for the graphics workstations). The only thing HDDs have to offer is capacity, and that's honestly irrelevant when project files are kept on a network share. The lowest cost HDD is only about $30 less than an SSD that will get the job done just as well (and several orders of magnitude faster). The accountants here have figured out that the increase in efficancy is well worth that little bit extra (but mabye I'm just lucky on that front). This data is also likely skewed by the fact that we build servers and workstations in-house (I need something to do between tickets after all), so we don't have to pay the eye-watering premiums for SSDs that OEMs are still charging.
How long until average home users have a similar setup with a shared network storage (attached to a router maybe) feeding files as needed to their PC, remote laptop, mobile, or tablet? This means there will still be a market for high capacity HDDs 5 years from now, but SSDs might ship at least 5-10 times as many units. Some say that Flash storage will be threatened by something even better by then, but I haven't seen a convincing demo of any such technology. Let's hope Western Digital has something cooking behind closed doors.
Paris, because she has a lot going on behind closed doors.