There are rumors swirling around, again, that IBM is looking to sell off all or a portion of its x86 server business to Chinese PC maker and server partner Lenovo Group. US channel mag CRN caught wind of something going down between IBM and Lenovo on Thursday and referenced an anonymous source with knowledge of the talks now …
IBM selling the PC client biz was a dire portent of what awaited PC sellers on the Vista launch. If this rumor proves true does it foreshadow an equivalent collapse of x86 server profitability? That would leave HP and Dell exactly where?
Re: Storm clouds
It has already happened. There is next to no profit in x86 servers. Aside from Cisco, who has an obvious advantage with their switches, no one is doing well. It doesn't surprise me that x86 profits are close to nothing, but it does surprise me that all of these companies are struggling at the same time. Dell, HP, IBM, Oracle/Sun... where are people buying their servers? Cisco has picked up a decent amount of market share, but it still seems like a small portion of the overall market. Clearly they have not all decided to go buy Unix servers instead. The global recessions seems to either be shutting down IT projects, or people are hanging on to their servers until they break, or the increased processing power of the new servers has caused all of the businesses to price themselves out of the market by not increasing price in line with processing improvement, or some combination of the above.
Re: Storm clouds
This is like comparing chalk to cheese..
IBM's problem is a simple one - their x86 and storage lineup are too complicated, try finding any information about their servers, build options etc, its next to impossible. They have effectively commited x86 suicide by failing to communicate with new and existing customers. As somebody else pointed out in another thread, this is probably intentional as they prefer the "service provider" rather than "vendor" model.
The real problem is, why don't they just come out and say their intentions. It would be better for customers, it would be better for Share holders, in the end it would also be better for the company.
Re: Storm clouds
Although the x86 market isn't an overly profitable one at the moment, there are other benefits for the likes of Dell and HP in remaining in it, less for for IBM which has already ditched personal systems. The buying power and economies of scale help prop up other platforms, like personal systems and storage. Many of the storage solutions on the marketplace are increasingly based on x86 server platforms, and maintaining control of design and manufacture of that is useful. Also, there is a lot to be said for the market presence and penetration that servers get you. I'd wager that HP and Dell sell a hell of a lot of other 'stuff' on the back of commodity server sales. I'm not being critical of IBM's strategy as I think it is also a valid one - the less valid approach is a half way approach where all benefits from either strategy are diluted so for IBM to increasingly get out of commodity markets is another move in the right direction.
Re: Storm clouds
The arse has fallen out of the physical server market due to Virtualization.
It doesn't make financial sense to spunk money on physical hardware. I'd wager IBM's storage business is reaping some of the rewards of this.
Re: Storm clouds
Rewards how? It would mean they have to buy in the hardware components for their XIV, V7000 and probably some other storage platforms, or develop proprietary platforms for them which increases cost and slows down development. And also a hell of a lot of commodity servers still get bought, and it's a lot easier to attach storage to a server sale than it is to compete with someone who is!
Re: Storm clouds
IBM is a deeply conservative 102 year old company. They have survived many technology transitions including the end of the typewriter. They look far ahead. They did a flub on the PC software thing, contracting their direct OS competitor to help them build their next-gen OS (OS/2 2.0) and it almost killed them, but they survived that too and won't make that mistake again. I have no idea what was going through that guy's head when he signed that deal. They turned into total jerks for a while, trying to take control of the language with that "planar board" and "fixed disk" nonsense. I think we're past that now. The attempt to jiujitsu yourself to dominance by taking control of the symbolset and thereby inspiring mass hatred should, though, be remembered as an object lesson. We never did find out who was responsible for that incredibly stupid idea. I'm fine with waiting until his estate publishes his memoirs; the pace of the PC being what it is the poor sap has probably still got mouths to feed.
They wisely got out of the client PC market as it was turning even more unprofitable than it had been for them that far on the eve of the Vista launch.
Now their market cap is more than Intel, AMD and the top-5 PC OEMs put together. They're dancing with passing Microsoft. I haven't been a big fan of their products since the early '80's but I have a deep abiding respect for their ability to run a business. They don't bob and weave with the fads, but they get 'er done.
I wouldn't want to work there though no matter how well it paid. Their whole motion makes me think their corporate structure could use more fiber.
Dave Cutler is one of the critical people in the evolution of software technology, currently a fellow at Microsoft. While doing background for this comment I discovered that the OS/2 wikipedia article has more interesting information on him than his personal and other related pages: http://en.wikipedia.org/wiki/OS/2
Basically he convinced Digital to not invent the PC, then skipped to Microsoft where he scuttled their IBM deal with huge delays, deriving NT from what he had learned at Digital and denying it to IBM who had contracted for OS development parallel with Microsoft's other efforts. Crucial moments in the evolution of modern tech. This was the dirty trick that finally broke the MS / IBM relationship, and vaulted Windows NT to the fore.
/What? You don't do background research for blog comments too?
Re: Storm clouds
Astute observation....IBM leads in moves like this...others follow...eventually ...or fade away
Mashing up Wikipedia articles with urban myths and reportage from partisan web sites is not background research.
Re: Storm clouds
Why should it be a surprise that ALL x86 server vendors are struggling when the only difference between them is the colour of the paint?
Re: Storm clouds
I wouldn't want to work [at IBM] though no matter how well it paid.
I worked for IBM a couple decades back thataway, and it was a lark.
Of course, I was working with a small group - first part of ACIS (academic computing) and then TCS (technical computing services, basically internal consulting) - in the same building as the Cambridge [Massachusetts] Scientific Center, where the culture was decidedly more relaxed than in, say, Kingston or Poughkeepsie. It was years before I even saw a copy of the IBM songbook, and no one ever asked me to sing anything from it.
Greybeards, and their female equivalents, ruled the CSC, and they didn't like rules that got in the way of their experiments. (Lynn Wheeler is moderately famous in some Usenet haunts for reminiscences of his work, and wife Anne's, at the CSC and similar IBM clubhouses.) Besuited managers and mid-level execs would stop by once in a while and suggest we should square our shoulders and march in step, but it never lasted after their visits.
My point is that IBM is a big organization, and not homogenous. Observing it from the outside may not give you an entirely accurate or detailed picture of what's on the inside.
 Of course there's an IBM songbook.
 One time some division VP or some such dropped by. At the time my officemate and I were working on software for a prototype graphics card. Our machines were named for artists, and our manager had purchased a couple of prints by the eponymous artists for us to hang in the office - a Rossetti and Matisse's Blue Nude for me, which of course was referred to as the "Big Blue Nude". The VP was horrified that a depiction of bare breasts, however acclaimed by art critics, adorned our office wall, and made us put Post-It notes over the BBN's naughty bits.
Re: Mikel Re: Storm clouds
Dell and hp have profitable PC and x86 server lines. IBM sold an unprofitable PC bizz to Lenovo, which Lenovo made profitable. What does that say about IBM management then and now?
Re: Mikel Storm clouds
HP Profitable? Last I checked HP did not have a P/E ratio because you actually need a profit. I looked it up -$6.55 per share. HP is a fool.
Re: IBM is a big organization
No, Apple is a mere company while IBM is a country
Re: Storm clouds
Well, IBM is now selling it's x86 server division because of bad margins. Earlier IBM sold the x86 PC business, for the same reason. The POWER division sees the same problem, margins is getting lower and lower. There will be a time when IBM has to sell POWER too, and only keep the high margin Mainframe business. Alas, IBM has been transforming to a service company for a long time now. IBM is exiting hardware, and selling hardware divisions. Not much left soon. The trajectory is going down. But of course, IBM profits as a service company, not as a vendor.
Re: Allison Park Re: Mikel Storm clouds
"HP Profitable?...." Nice dodge of the fact the PC part is. I also note you do not deny that Lenovo took a loss-making IBM PC business and made a success of it.
>Basically he convinced Digital to not invent the PC
I think you meant he helped to convince Ken Olsen not repackage the MicroVax as a desktop workstation in the PC marketplace and price bracket.
Funny how in the mid 80's DEC were in the position to: launch a VMS-based workstation (with an extensive third-party applications catalogue) and release VMS to X/Open (now The Open Group) to become the standard for POSIX, and the rest as they say would of been history... Instead they left the PC to Microsoft and IBM and X/Open basically adopted Unix to become the basis for POSIX.
Re: Allison Park Mikel Storm clouds
not really a dodge.... just a bigger picture view.
Lenovo has done very well. I never viewed IBM as a consumer company. From looking at their acquisitions and divestitures it sure looks like they want to get out of commodity (hard drives, point of sale) and retain high value. I would not view this as a hardware general issue and I think they make higher margins on hw than services but sw eclipses all.
should be interesting either way i think customers are happy with ibm and lenovo can't say the same for my friends watching the HP ship burn
Re: Allison Park Re: Allison Park Mikel Storm clouds
"....and retain high value....." That's a very convenient way to describe IBM's retreat into the high-end, having been soundly thrashed by Dell and hp yet again. But you forgot to mention that IBM's high-end, especially the mainframe biz, is also trending down - where do they retreat to when mainframe becomes too unprofitable?
What IBM are doing is cost-cutting and share boosting by sale. It's a neat little scam - take a failing business unit, sell it to Lenovo (whom IBM own a chunk of, so IBM end up paying themselves) and add some billions to the year's figures (looks good on Wall Street), and then use that to buy back some shares and boost the overall share price, keeping the thicker shareholders happy. Sooner or later the IBM board will run out of family silver to sell, and that is when the shareholders will wake up and realise they have been scammed.
I seem to recall that there have been some sort of warnings recently about buying laptops from Lenovo (and/or other China-owned companies) and have to think that there would be similar warnings about servers from China-owned sources.
So I would think that if IBM sells their x86 server business to Lenovo, then the value would fall and possibly fall precipitously, as at least some companies would be wary of buying servers from a China-owned company, I should think.
Not happy about that, but I guess it was inevitable.
Lenovo has been selling a "ThinkServer" line for quite a while, which are basically clones of the very low end System x towers and rack systems. I think it was branded as some kind of cross licensing deal, but I guess we know what was really going on.
Hopefully Lenovo will keep the design and engineering teams in North Carolina -- they're still working on most of the business-grade hardware designs for the PC business, and the System x/BladeCenter guys are there too, so it's a good fit.
I guess the only good thing is that even with the recent cheapening of the designs, the ThinkPads and ThinkCentres do retain a good chunk of the original IBM design...they're still my favorite laptop vendor. So hopefully, if Lenovo is smart, they won't mess around with quality too much. IBM's System x gear is (was?) top quality, easily as good as or better than the HP ProLiant. And at the high end of the range, their hardware designs are really interesting. The downside is that it's very expensive compared to Dell, slightly more than HP. For that price difference though, you get US-based tech support who really understands their stuff. I've never had a problem with IBM System x and BladeCenter support (It's a dedicated group sitting in Atlanta if you're in the US,) but have wanted to reach through the phone and strangle HP and Dell "support" representatives.
The fact that IBM is keeping FlexSystem, System z and System p in house seems to me like they're betting that everything will collapse into the cloud, either public or private, and no one will need physical x86 boxes anymore. FlexSystem is basically BladeCenter 2.0 + Cloud-in-a-box, and competes with the VCE stuff. It looks neat, but just like the VCE stuff, I can't afford it to play around with in the lab. However, I wonder what's going to happen with OpenStack gaining popularity now -- people won't be locked into VMWare or IBM and they'll actually want cheap x86 boxes that they can turn into a loose collection of VM containers. Plus, you have places like the one I work in, with highly distributed branches and crappy network connectivity all over the universe that can't do the consolidation thing cost-effectively who will be buying physical boxes for some time to come.
This is kind of a bummer to me though -- I live in New York, and IBM still has a fairly big presence here, since their HQ and main research lab are in Westchester and Dutchess counties respectively. Just like the rest of the US though, that presence keeps dropping with every physical product line they sell off. I don't wish ill on other countries, but part of me is hoping for a Chinese economic collapse just so MBA management consultants will stop telling executives that you can't manufacture physical products, for any price. Lack of physical products or selling that off to the lowest bidder means no long-term hardware innovation, and we're going to end up stuck in a rut forever. (I was a rust belt kid and watched all the manufacturing move to the South, then overseas. Not fun.)
Re: Not happy about that, but I guess it was inevitable.
Lenovo semi-recently announced that they are expanding their server HW presence in NC, including bringing some manufacturing over from China. Coincidence?
Re: Lin78 Re: Not happy about that, but I guess it was inevitable.
"Lenovo semi-recently announced that they are expanding their server HW presence in NC, including bringing some manufacturing over from China. Coincidence?" I suspect it will be enough local assembly for them to slap a "made in America" sticker on the product (and qualify for local grants and tax relief).
With the anti-chinese paranoia running through USG at the moment (whether justified or not), can't see certain lobby groups being too happy at this.
Given the nosedive in Quality between IBM ThinkPads of old, and the newer Lenovo machines, I would not consider a Lenovo server for any enterprise application.
Their T-series are still OK -- I don't agree with the latest design changes in the keyboard and trackpad, but HP's business laptops are fugly now. Not that the ThinkPad was ever sexy, but it's still a great Spartan business tool.
Hopefully Lenovo will stop pinching pennies on the T-series design, or pull an Apple and go back to high-end high-margin IBM-style design. These days, with PCs lasting way longer than even the 2000s, you have to build some obsolescence into your products or no one will buy new ones. But, if you charge $2500 for a laptop, and the user is happy for 5 or 6 years, you win anyway. I'm not a huge Apple fan, but I do have a 2008 MacBook Pro as well that has been through almost the same torture as my T-series ThinkPads. If you pay for it, you get quality. If you pay $299 for a laptop from Best Buy, don't be shocked when it dies randomly.
Not that the ThinkPad was ever sexy, but it's still a great Spartan business tool.
Agreed! I use mine to keep track of my Helots and organize my phalanx.
I like to keep track of my Harlots :)
I'm on my 3rd T series because the fans keep failing, seems to be a well known defect judging from the Lenovo forums.
Yet my old Thinkpad 380Z just soldiers on as an old game station.
I think every major vendor will adopt the Oracle model. Dump "commodity" give all profit to Intel and move into premium high value systems which give less importance to intel / power / sparc
lenovo seems like the right choice but you never know about samsung, fujitsu or hitachi
Re: not surprised
Perhaps we are assuming x86 has a future anyway, I'm not sure it does. Intel has managed to fab ist way up market for years, with a pretty poor architecture, but pretty soon they will see the limits of physics. On the other hand ARMs will be coming from Calxeda, via Dell, HP et al, and they will compete well versus x86 for web servuces. So IBM becoming an ARM licensee, flogging off the x86 business for $5B might well be a very clever move.
Re: not surprised
"I think every major vendor will adopt the Oracle model. Dump "commodity" give all profit to Intel and move into premium high value systems which give less importance to intel / power / sparc"
Agree that no company wants to be in the business of pushing 2U x86 servers with no value add. All of that business will eventually go to the Lenovos, Acers, etc of the world who will make them very inexpensively, if not particularly well. I think IBM and, to a lesser extent, Oracle still want to be in the CPU business, but they want to be in their own CPU business instead of just putting a case around Wintel. CPU will not be the central part of their value proposition, software will be, but they will be another level of differentiation. System I/O and board design is generally more important than having a CPU that is way better than Intel. Most people run out of memory and I/O long before they tap out CPU.
Re: not surprised
"Perhaps we are assuming x86 has a future anyway, I'm not sure it does"
Agree on x86 vs ARM, but I don't think that does much for IBM and the other OEMs. Instead of commoditized x86 servers, the market will just shift to commoditized ARM servers... and the same problem exists for the OEMs. Everyone is using the same chips, software, etc without much differentiation.
Re: not surprised
If you have dealt with Oracle support you would really not want to go there!
As for the non-x86 market, I can see ARM going places because it is better for power/performance than x86 (for a given fab technology, but Intel currently has a ~18month advantage there) and it is not vendor-specific. Most folk I know don't want to be in a position where they have to keep buying high priced hardware from one vendor.
If IBM, etc, want to play that game exclusively then the system has to be *VERY* much better than x86/ARM + Linux/Windows commodity solutions.
okay stop laughing, IBM has to.
not all x86
its just the low end stuff, think 3650 and below plus towers, all the 2 socket boxes. quad, blades, P, Pure, and Z all stay in house. No way the US Feds would allow P/Z biz to go to the chicoms. keep in mind, IBM owns like 60% of Lenovo.
More layoffs in the offing. IIRC, revenues in x86 servers are still rising but the margins are such that a company the scale of IBM cannot seem to reduce their overheads enough. Lenovo turned the ailing PC Division into a profitable business and the only way they're going to be able to do this with wintel servers is by taking the same approach: 1) Turn them into true commodities like toasters and 2) lay off those expensive European and American workers and replace them with any number of dirt-cheap Chinese.
Increase server market share
So Lenovo would take on and keep the IBM patent research, design and engineering teams in Raleigh NC like they did for the Thinkpad/PC range in 2004. I read recently that Lenovo have an 11% increased market share, while the rest including Dell, Tosh and HP have all fallen.
If it happened, and if IBM and Lenovo get the right message out there, loyal IBM System x customers are likely to transition over to the same System x boxes/ roadmap / but with a Lenovo badge. Lenovo would I assume, also take on selling all the IBM system x options and spares. With lower manufacturing costs and pricing, we could then see the Lenovo owned system-x range start to take market share from HP & Dell like they have achieved in the desktop/laptop marketplace.
Not forgetting that the high-end IBM eX5 system x intel server architecture range is not a commodity product and has unmatched memory expansion when compared to HP and Dell rival products.
- Game Theory The agony and ecstasy of SteamOS: WHERE ARE MY GAMES?
- Intel's Raspberry Pi rival Galileo can now run Windows
- Hello, police, El Reg here. Are we a bunch of terrorists now?
- Microsoft and HTC are M8s again: New One mobe sports WinPhone
- Worstall on Wednesday Wall Street woes: Oh noes, tech titans aren't using bankers