Areal density matters
What is slightly concerning at Hitachi GST is that it might just have dropped behind the industry leading edge with regard to areal density improvements. It has only just announced its 2TB drive, the Deskstar 7K2000, whereas Western Digital and Seagate brought theirs out in the first half of 2008. Also, and tellingly, the HGST 2TB drive needs five platters to do what four do for its competitors. Admittedly, this enables a faster 7200rpm spin speed than the industry leaders, but there is always a cost to using an extra platter and head.
Oddly, Hitachi GST did announce a refreshed 1TB drive, the Deskstar 7K1000.C, which needs just two platters and so uses the 500GB/platter density that would be needed for a 4-platter 2TB drive. What is going on? The 7K1000.C is not yet available, which suggests that the 500GB/platter media or heads were just not ready when hardware decisions were made at cut-off dates in the 7K2000 development process. Alternatively there are manufacturing capacity issues and HGST felt the 500GB platters would be better used in the 1TB drive and not the 2TB one. It seems that something is not quite kosher here.
The thing is that, according to head manufacturer TDK Corporation's roadmap, the 3.5-inch drive areal density is jumping up from 500GB/platter to 640GB/platter, meaning we could see 4-platter 2.5TB drives. WD and Seagate could conceivably announce theirs in April, a year or so after their first 2TB drives were introduced. But Hitachi GST has only just introduced its 2TB drive and its first 500GB/platter 1TB drive. To announce a 2.5TB one and use 640GB/platter technology in April would mean that it would get no good return at all from its 500GB/platter tech.
There could be a nine month or longer lag opening up here.
SFF drive areal density
Areal densities on its 2.5-inch small form factor (SFF) drives appear to be suffering too. Hitachi GST announced a 10,000rpm 300GB 2.5-inch drive, the UltrastarC10K300, in March this year, with a 6Gbit/s SAS interface. Late last year it had a 500GB small form factor drive with a 375Gbit/sq in areal density. But it has not yet introduced a 640GB SFF drive, as Seagate has with its 640GB FreeAgent Go.
Western Digital has a 3-platter 1TB SFF SCorpio Blue drive with a 333gB/platter areal density. That seems to be on the SFF areal density leading edge, and TDK's roadmap confirms it, with a transition from 250GB/platter to 320GB/platter occurring now. Will Hitachi GST introduce a 320GB/platter SFF drive by the end of the year?
If it fails to keep up with the leading edge of areal density development then it will cede retail shelf space and OEM market share to WD and Seagate. Because Seagate let WD get ahead in SFF disk areal density last year, it now faces the possibility of being overtaken and relegated to the number two drive manufacturer position in terms of unit shipments. HDD manufacturing businesses are like massive super-tankers. Changing course takes ages and speeding them up and slowing them down happens at near-glacial speed.
HGST has virtually no chance of catching up with the two industry leaders unless it maintains areal density equality with them or, even better, gets an edge.
The 7K2000 and 7K1000.C areal density pattern suggest that it might be falling behind the curve in this essential parameter of HDD manufacturing success. A 320/333GB/platter SFF drive introduction would quell such concerns. It could be that Hitachi GST is coming towards a tipping point. Staying up with its competitors on areal density improvements will ensure it has table stakes for competition with them. Falling behind will mean that it will likely face an anxious time during which it will try and catch up while seeing relatively diminished revenues and higher costs.
Competition is good, so let's hope that Hitachi GST does the business and keeps up with the leaders. That will spur them on to build better products and help all of us get more disk drive bangs for our bucks. ®
Could HGST be going to the wall? What a shame!
Oh wait, no, that was sarcasm. What I meant to say is good riddance to bad rubbish, and don't let the millions of dead drives hit you on the way out.
The Deathstars were bad enough, but it was their idea of customer service (aka: putting sand in the Vaseline™) that really nailed their coffin closed IMO.
Terrible company, terrible service, and terrible products. You won't be missed.
it's back to front, no?
They bought the HD manufacturing facilities from IBM, but not the experts or the history in R&D, and they're flopping.
Big surprise? Uh, no. Sharp people are more important than good facilities.