Makes sense. In the era of cloud services IBM's traditional targets are going to be buying far fewer servers. And you know Amazon and Rackspace aren't buying IBM hardware. Name-brand computers are a shrinking business. This is true across all markets. Even Apple's share is shrinking (as it continues printing money with the devices it's selling, of course)
I would not be surprised to learn that the US Government is the largest single purchaser of IBM hardware, directly or indirectly. Most projects I am familiar with that use COTS machines for processing are using IBM boxes. A significant number of those were x86 though, which leads me to believe that market isn't enough for IBM. Of course, they were pushing POWER really hard when I left that world.
You may want to read GP's comment again. He believes that all branded computers are on their way out and he uses the (fictitious) decline of Apple's PC business as proof. GP is obviously not talking about just Apple's server offerings. It'd be pretty silly if he were, since Apple has never had any meaningful marketshare in servers.
FWIW, a Mac mini server is a regular mini with two 2.5" hard drives (not hot-swappable or even user accessible) and a token to download Server.app from the App Store. Apple can't even be bothered to list a rack mount kit or 220V PDU-compatible power cable in their store.
Apple has never had a meaningful market share in the server arena. It's virtually 0%. It can't shrink because it doesn't exist in the first place.
Xserve and XSAN were never specifically for enterprise, but rather as support infrastructure for things like large Avid or Final Cut installations. There have been only a handful of cases where Xserve was deployed en-masse.
I don't think they would as most of the Power stuff is linked to AIX licenses and is fairly "unique".
The offload of the x-series business is most likely due to the fact that IBM can't compete on price with the volumes which HP & Dell shift. IBM have no real USP on X86 as enterprise are now mostly putting ESX, Xen or Openstack on them, IBM have no potential license or service fees generation.
It doesn't make it a stupid decision. Levano can already take the low end (or Asus will). IBM sells their low-end while it's still worth something, rather than fighting a losing battle.
Then lenovo puts in a few coins into R&D and takes the market slightly up from the low end. Which IBM gladly gives away because the margins aren't high enough...
Then Lenovo puts a few more coins into R&D and takes the midrange. Which IBM gladly gives up because the margins aren't high enough...
This has happened more than a few times across several industries - American cars, Japanese electronics, mainframe/mini/microcomputers, etc.
EDIT: Also see apricot computer. Which did the /exact/ same thing - outsourced the production of it's "commodity" computers, losing its special sauce, and lucrative service contracts. It was gone in a flash. Sir John Harvey-Jones even did an episode of the UK show "the troubleshooter" on them. It's a sure way to make a quick buck this quarter, but horrible long term.
IBM does a little to enhance their x86 line, but I don't think it's of any great value to many customers.
If they were making a lot of revenue (not even profits!) off their value-adding, and this value-adding could be applied to their top-end stuff (thus amortised across a larger base), it would make sense.
I'm not an expert in IBM servers, but I don't think they are getting a lot of revenue at the bottom-end as a result of their bells and whistles, and it's probably not that easy to transfer those enhancements to their top-end servers.
And replaced them with the knowledge economy right. Which has some structural issues right now across the developed world i.e record corporate profits, less than record employment rates.
Why would POWER be affected by this decision?
Everyone and their mother sells a x86 server. Hard to make a profit when everyone is competing. How many other companies sell POWER hardware?
While from a business standpoint, this is perfectly viable and a smart move on IBM's part, in the long-term, seeing how Lenovo has ruined the Think brand, I would be wary of spinning off any more operations to Lenovo.
They just don't have clear product focus, but in this world, I guess shareholders rule. Even if those shareholders don't act in the best interest of the company.
This actually looks like a great laptop, could some point out what's wrong with it?
The screen isn't too big and it has a nice resolution, which isn't easy to find.
Many people are worried about the switch to a chicklet-style keyboard. I've been very happy with my previous employer-provided Thinkpad T-series laptops and purchased an X220 online, knowing the keyboard would be great. I wouldn't order one of these new ones without trying it first.
First of all, there's no seventh row on the keyboard. Second, the Trackpoint buttons are gone. Third, the Thinklight has been replaced with a backlit keyboard, which is egregious as the Thinklight allowed one to illuminate documents or other materials as well as the keyboard.
It's just like any other laptop now. And when you're used to 1440x900 (I'm on a MacBook Air) 160 horizontal pixels is really worthless and screws up the proportions. What ThinkPad users want, and have wanted for ages is a 16:10 display, excellent engineering, and the TrackPoint front and center.
I've had two Thinkpads in the past (A21 and A31) before I switched to a Macbook Pro. I don't remember the thinklight being very useful, but that may just be me. Not sure how well the trackpad works as a mouse button.
It's a shame 16:10 has not been adopted by more laptop manufactures, it's a great resolution. Even 4:3 may be ready for a comeback, I'd really like to see a Chromebook in action, although I would prefer to run a more traditional Linux operating system on it.
I remember when they spun/sold off a lot of their peripheral production to Lexmark. Now their systems are going to Lenovo... I think IBM is really following Microsoft/Oracle... interesting that this comes as the likes of Oracle and MS are losing mindshare regarding up and coming developers.
This was bound to happen once servers all standardized on one architecture. Not that standardization isn't a good thing, it just became a race to the bottom just like the desktop market.
I'm kind of surprised, actually. There are still plenty of ways to differentiate your product in the server space, particularly when it comes to high availability. My company (Probably in the top 20 in the US in terms of revenue) pays outrageous amounts of money for servers because that's what it costs to fill your data center with DC powered high availability hardware.
Perhaps they want to sell the x86 business and focus on the emerging ARM/alternative architecture business. I wonder what sort of non-compete the deal would include.
Being that IBM's POWER architecture is also a RISC architecture I would expect them to rally behind that, if anything. I've heard that POWER's performance had improved dramatically recently, so it would not surprise me if they were trying to push POWER.
The article quotes an analyst at the end: "IBM's troubles are indicative of bad news for the whole technology business".
A terribly poor & overreaching conclusion. It's an odd quote to single out in your conclusion too. The x86 server market is actually growing. What's down is IBM's share of that and more importantly the margin they're able to extract.
What is happening is that it's getting so big that it's increasingly commoditized. People are still buying hardware.
IBM is the biggest player in the x86 server market, but their share and margin is in decline. They still make great margins & growth on z-series.
44 comments
[ 3.4 ms ] story [ 120 ms ] threadhttp://en.wikipedia.org/wiki/Xserve
"On November 5, 2010, Apple announced that the Xserve line would be discontinued on January 31, 2011"
http://www.everymac.com/systems/apple/xserve/specs/xserve-in...
No, it’s not. It has outperformed the rest of the PC market for years and continues to gain marketshare.
http://www.asymco.com/2013/04/11/happy-birthday-ipad/
The context is server hardware. Apple gave up on the market; the only relevant offering is the very low end Mac Mini Server.
Also, Apple doesn’t only sell Mac Mini servers, it also sells Mac Pro Servers. Priced at $3k-12.5k, they’re hardly low end. http://store.apple.com/us_smb_78313/browse/home/shop_mac/fam...
That said, they work well enough.
This completely ruins your comment. Please try to keep things friendly.
Xserve and XSAN were never specifically for enterprise, but rather as support infrastructure for things like large Avid or Final Cut installations. There have been only a handful of cases where Xserve was deployed en-masse.
The offload of the x-series business is most likely due to the fact that IBM can't compete on price with the volumes which HP & Dell shift. IBM have no real USP on X86 as enterprise are now mostly putting ESX, Xen or Openstack on them, IBM have no potential license or service fees generation.
It seems a sound business decision to me.
Didn't the american car companies take that route too? And steel mills. and... mainframes I think.
Then Lenovo puts a few more coins into R&D and takes the midrange. Which IBM gladly gives up because the margins aren't high enough...
This has happened more than a few times across several industries - American cars, Japanese electronics, mainframe/mini/microcomputers, etc.
EDIT: Also see apricot computer. Which did the /exact/ same thing - outsourced the production of it's "commodity" computers, losing its special sauce, and lucrative service contracts. It was gone in a flash. Sir John Harvey-Jones even did an episode of the UK show "the troubleshooter" on them. It's a sure way to make a quick buck this quarter, but horrible long term.
If they were making a lot of revenue (not even profits!) off their value-adding, and this value-adding could be applied to their top-end stuff (thus amortised across a larger base), it would make sense.
I'm not an expert in IBM servers, but I don't think they are getting a lot of revenue at the bottom-end as a result of their bells and whistles, and it's probably not that easy to transfer those enhancements to their top-end servers.
They just don't have clear product focus, but in this world, I guess shareholders rule. Even if those shareholders don't act in the best interest of the company.
Did they really?
It's just like any other laptop now. And when you're used to 1440x900 (I'm on a MacBook Air) 160 horizontal pixels is really worthless and screws up the proportions. What ThinkPad users want, and have wanted for ages is a 16:10 display, excellent engineering, and the TrackPoint front and center.
Lenovo is squandering all of the above.
I've had two Thinkpads in the past (A21 and A31) before I switched to a Macbook Pro. I don't remember the thinklight being very useful, but that may just be me. Not sure how well the trackpad works as a mouse button.
It's a shame 16:10 has not been adopted by more laptop manufactures, it's a great resolution. Even 4:3 may be ready for a comeback, I'd really like to see a Chromebook in action, although I would prefer to run a more traditional Linux operating system on it.
I don't think we ever buy from IBM, though.
edit: Yes, I did mean 'rally' - thank you.
A terribly poor & overreaching conclusion. It's an odd quote to single out in your conclusion too. The x86 server market is actually growing. What's down is IBM's share of that and more importantly the margin they're able to extract.
What is happening is that it's getting so big that it's increasingly commoditized. People are still buying hardware.
IBM is the biggest player in the x86 server market, but their share and margin is in decline. They still make great margins & growth on z-series.