All I know is the WD has bought the HDD division from Hitachi....
And Seagate bought LaCie and Samsung.... Which means there are really only two left.
But, regarding capacities, I suspect the real answer is something like this:
* Over the past few years, platter density improvements and SSD-based caching systems meant that Data Center buyers are no longer very interested in paying top dollar for high-performing 10k RPM SAS drives. While they got an overall price bump from the flood, Enterprise buying has been dramatically shifting towards the value segment. Why pay top dollar for a high performing, reliable SAS drive? Just buy a bunch of cheap, huge "green" drives (which has the bonus that you can call it part of a power-saving initiative) and stick them behind a SSD cache with a huge DRAM cache (oh yeah, DRAM prices plummeted too).
* Conveniently, many of the newfangled RAID-like systems provided more economical and convenient redundancy protections. These systems have improved and gotten cheaper, and have become much more widespread (driven by vendors like NetApp), so there is less of a motivation for Enterprise buyers to spend extra on "RAID-certified" class drives for reliability.
* The demand "portfolio" has been slowly shifting away from consumer-driven and towards enterprise-driven.* PC sales are down everywhere, again (except Apple), and much of the rest of high-end consumer use has shifted to SSDs. SSD manufacturers took advantage of the spike in magnetic storage costs to dramatically cut prices on (and spur consumer adoption of) NAND-based storage solutions. There's no longer any money at all in the lower capacity drives (as what wasn't eaten by SSDs raced to the bottom).
* On the other hand, data center storage needs are growing faster than ever. Suddenly, filling all those Dell PCs is a much smaller slice of your business than filling all of those racks at Facebook and Twitter and Amazon E3.
Then...
* Platter density developments stalled somewhat (or were set back) due to the flood.
* While there
was a real production problem from the flood, WD and Seagate's dominant positions (and clout with suppliers) allowed them to suck up most of the remaining capacity. In other words, it was really tough to get parts, but WD and Seagate were able to ensure that they received the lion's share of what
was available. This hurt the smaller players (and they weren't able to profit as much from the price gouging windfall).
* The consumer-to-enterprise shift, combined with the real supply constraint from the flood, allowed the two "winners" to get
and maintain higher margins for their high-capacity drives for the first time... Ever, probably.
* These higher margins not only taught both WD and Seagate a lesson, it gave the biggest two players suddenly large cash reserves that they could use to consolidate the market (thereby helping to preserve the higher margins for longer).
So... In the recent past, WD and Seagate and HGST and Samsung Storage were in a "race" to continually bring out the new "biggest drive". Because, the biggest drives were the
only drives that commanded big margins. Soon after the newfangled 4TB drive came out with the higher platter densities, the bottom would fall out on the lower-capacity drives. Sure, they'd keep selling the higher-margin 7200 RPM "Black" and "RAID" drives, but no one was buying them in massive numbers anymore, so they were a much smaller percent of the bottom line and couldn't hold overall margins up anymore.
As an example, I bought two 3TB "Green" drives just before the flood for $119 shipped. That price had been steadily, but quickly, falling since the initial 4TB drives came out. So, the vendors are motivated to ship the highest capacity drives they can.
Now, they've consolidated, and the margins on those 2TB drives are holding steady. The new platters aren't coming online as quickly as they would have because of the flood, but there's no rush. Why put out a 5 platter drive to hit that 4TB number now, when you can sell two 2TB drives for more money? And, if you aren't "wasting" margin on your precious top-tier drive by including a metric ton of platters, there's no drive to push on the engineering nerds to accelerate the platter density jump (so that you can get that sweet couple of quarters where you're now selling a 4 or 3 platter drive for the same price/capacity that you were with a 5-platter drive). You stop calling their managers demanding progress every five days, and they let their workers use vacation days and whatnot.
Plus... Like I said, they learned a lesson, and there was a massive consolidation in the market. We have an effective duopoly now. Their only real threat is NAND, and so long as the "other guy" doesn't mess with you too much, you can both sit and take your time and actually make some nice money on the top capacity drives.
So, those higher capacity platters will come, and the 4TB drives will become the new 2TB drives (and so on and so forth), but they're just not in the rush that they were 18 months ago. NAND is way off from threatening the high-capacity part of the game, and with fewer competitors, the game got way easier to play.
* This is why we saw Warranty length plunge over the past few years. The shift from Consumer-driven to Enterprise-driven market, meant that the companies were motivated to lower warranty coverage on their "consumer" drives, in order to prop-up demand for the enterprise-focused ones (buying departments often love warranties). But, the manufacturers had long since gotten completely non-reality based with their MTBF numbers, so enterprise users no longer believed them that the high-priced "RAID-class" drives really were better than the consumer cheapo drives. So, we have enterprises saying "Who cares if it dies. Buy two of the cheap ones instead. Replace them when they die." (The better modern RAID-like systems give them this luxury.)
Author
Topic: what's up with 4TB drives? (Read 28031 times)