Seagate defrags 14% of workforce: 6,500 axed Seagate is cutting 6,500 staff, or 14 per cent of its 46,000-strong workforce, across Asia, Europe, the Middle East and the Americas. Those losing their jobs will shown the door over the next 12 months; the redundancies will cost the hard drive maker US$164m. These layoffs are on top of the 1,600 heads put at risk at the end …
Crucial and other Flash makers caught it. I doubt Seagate will come back from this, and if they truly believe that the need for mechanical drives will only increase, they're either delusional, stupid, or simply lying to their investors. I expect a slow slide into irrelevancy with more layoffs, punctuated by scrambling attempts to introduce half-baked new products. Spinning rust isn't dead yet, but it's going the way of the DVD market.
"Conventional hard drives still have their uses. But as the size and cost of SSD increase and decrease respectively they will be relegated to specialist applications. They will be replaced by SDDs."
Agreed. Vacuum tubes still have their uses too, but only for speciality applications. Kind of a bad analogy I suppose, as audiophiles have a point, but I can't think of much that mechanical drives do better that will be relevant when capacities and costs for Flash catch up. Maybe they're better for data recovery after catastrophes. Most likely there will soon be another technology for mass storage, which will make spinning rust even more irrelevant. Maybe if Seagate's lucky they will get into that.
If you're into drive data recovery, you already lost. It costs a fortune and is by no means guaranteed. Ask my employer, and the reason why I have the previous guy's job (he didn't back up, they lost the main RAID, the disks could only be recovered by specialist recovery, but all they could do is stitch the RAID format back, not magically resurrect the dead disks - they lost a lot).
And magnetic history is a myth that nobody has ever demonstrated. As such getting data back from an SSD is likely to be the same, if not better, than from HDD. At least with SSD they *can* be designed to warn you, to lock into read-only (whether major brands do or not is another matter), etc.
But, really, that's not even a selling point. If you sold a hard drive that said "This drive is 5% more recoverable by specialists in the case of utter drive failure", you couldn't charge anything significantly more for it. People either expect them to fail or not, depending on the usage. And by that time, your backup and redundancy procedures have totally failed and you're into thousands of pounds of expense to get one drive back.
I literally cannot think of a single thing that an ordinary-priced SSD of decent size could not win, hands-down. And the more production capacity pulled from hard drives and put into SSDs, the cheaper they will get, relatively speaking. Hell, even the density is ridiculous. I opened up an SSD recently and it was 2.5" for a start, and then inside the metal casing (which was basically just two bits of metal "popped" together) was a single board covering less than 1/3 of the surface area of even a laptop drive.
If they had wanted to, and you could afford it, they could combine multiples of those boards onto one drive and in the space of a single desktop 3.5" drive, you could fit dozens of terabytes quite easily.
The only problem is cost. Throwing their money away on HDD production is going to be their downfall. Everyone else is dialling down everything but enterprise HDDs (which just take longer to get out of the market) and ramping up SSD production. No rare gases (helium is significantly rare, nowadays, yes?), no hermetic seals (literally two bits of metal casing snapped loosely together), no high speed drive dynamics, no tiny-micron-level moving parts, no fragility of shipping.
Sorry, but any manufacturer not planning for an exclusively SSD future (until the next tech comes along) is really just burying their head in the sand.
"Fragging", at least in the Vietnam War era, referred to doing away with a troublesome member of your own military unit IIRC. "Defragging" OTOH refers to rearranging bods (or data) to coalesce the free space into a larger block with any doing away with data.
It couldn't be because Seagate decided to cheap out on everything, and make crappy drives that have the highest failure rate in the industry by far, such that they are banned from my datacenter?
Storage vendors noticed too, all replacements I get in are *never* Seagate drives, replacing the failed Seagate disks by the dozens.
Indeed, and same goes for WD that introduced "features" on its "green" such as parking head every 8 second of inactivity which make those disks fail on Linux under 3 month. Worst they did that too on the "red" that are meant for NAS which obviously are 99,99% Linux!
And the prices still didn't went down where they were before the "great flood"... they always have an excuse like the dollar is high! In the same amount of time, SSD prices have been divided by 3 or 4.
I believe the utility name is WDIDLE3.EXE (at least for running from a freeDOS boot stick).
Even many of the RED drives were apparently wrongly set with a short idle timeout from the factory, as became apparent when they started failing after less than a year's use. The WDIDLE3 utility can report the current idle setting on a drive (via command line switches), so it's worth using it to check both 2.5" and 3.5" WD drives you may care about.
I can say that Seagate's "Momentus Thin" series of laptop drives is one of the biggest (actually, tiny and thin) pieces of crap I've ever had the misfortune to deal with. Second perhaps only to WD's "black" series. (which must be either the reliability or speed rating, or both)
Seagate Thin drives are based on the same model that the old Maxtor (remember them?) drives used. And were total rubbish then too.
However, I can't say I've much bad to say about WD drives generally, excluding the 2TB greens which were utter garbage and all epically failed. ALL of them.
I suspect Seagate'll come out with an OK Cupid-type app that locates the nearest defective drive of the same model as you have, asks both of you to walk the half block or so, meet up, and swap bad drives betwixt.
"Same procedure as now," Seagate can brag, "except quicker and you save on shipping charges!"
"Also, we no longer need the 6,600 people we've been paying to do that".
I think that the future of storage is less about whether you have spinning rust or solid state. Its about how you manage your storage - horses for courses.
The drive manufactures have been around a long time. They ought to have guessed where things were heading and use their current profitability to build their business for the next stage.
But they didn't did. Outsource the bulk of business to (cheap) Asia, dumb production, no R+D.
Most people's storage can be roughly divided into video and everything else. Video will always provide a market for cheap commodity performance, everything else will go solid-state with the big cloud vendors operating a tiered system for a while.
Spinning disks will have a market for as long as they can remain cheaper in TCO than solid-state.
Bulk Storage is still way way way cheaper on spinning drives right now.
Basic example;
£274.79 per terabyte (Crucal MX200 based) - as a middle of the road option.
£40 odd for a 1TB 7200rpm Drive with a terabyte.
Scale up to 6TB drives and the £24 per terabyte mark is common (in the non-server grade world as per the non-server grade SSD compared against).
£24 x 6 = £144.00
There is a long way to go before we'll see bulk storage move to SSD - as a good example of where spinning drives are going to last a good time yet.
However, I'm interested to know what sort of roles those Seagate people had - are these technical people? Sales people? Just how many people do you actually need to have drives made these days - isn't most of it robotic?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
