When you live in a glass house – like IBM does, both metaphorically and literally – you don't throw stones. So when Oracle suddenly pulled the plug on future software development for the Itanium processor back in March, Big Blue didn't say squat about the situation. But the company's top brass have been taking pot-shots at the …
There's H/W costs -that you can't sell below- and software costs, that aren't so clearcut. the R&D spend is already gone, and if you bill for support, that's paid for.
The reason IBM can undercut HP here is not the hardware, it's that Oracle costs even more than DB2
Replace proprietary hardware X and software stack Y with proprietary hardware A and software stack B
How is this breaking free again? You're still tied to one company and one software stack. Moving from one gaol to another.
In the IBM deal 16 cores is somewhat (not necessarily half) less powerful than the 32 it's comparing against. Also I'd have thought that DB choices are very sticky and that there'd be uproar within any company that said "we're moving DB 'cos the PHB got a deal on some IBM hardware".
Apoligies to John Deacon / Freddie Mercury
I want to break-free
I want to break-free...
I want to break-free from your lies
You're so self-satisfied I don't need you
I got to to break-free
God knows, God knows I want to break-free
"...about 391 of those deals came out of Oracle/Sun accounts..."
Next week: Larry announces IBM's "secret plans" to sunset the Power architecture and ceases support for Oracle on same.
RE: "...about 391 of those deals came out of Oracle/Sun accounts..."
At that rate of replacement, I don't think either Leo or Larry will be exactly breaking into a sweat! Take a look at the figures hidden in between TPM's Big Blue pompom waving and it doesn't exactly paint a picture of great success. The hp competitive conversions don't exactly sound like a tsunami of Power:
"....IBM says that it has identified over 10,000 HP Integrity server accounts.... " So that's the overall target. I'm sure it's no coincidence it's the same as the number of shared Oracle-Integrity accounts hp mentions in its court case against Oracle.
"....the first quarter of this year was pretty good, with 845 competitive replacements...." So this quarter just gone was extra good for IBM? So how many Integrity implementations could they have replaced with Power, assuming all those were Integrity and not old Compaq Alpha?
".....40 per cent came from HP accounts...." So, a grand total of 82. Ignoring the fact that IBM marketting probably counted individual servers rather than customers, and that some of those were probably very old hp kit rather than Superdomes, 82 is still nothing more than market churn. It's 0.0082% of the target accounts, or 0.033% if they maintain that over the year. At that rate, we'll all be flying Jetson-style aircars before IBM even makes a noticeable dent!
".....the kind of thing Big Blue has been doing for years and presumably HP and Oracle/Sun are doing, too, but they don't talk about it (which is idiotic if they are doing lots of competitive takeouts)....." I'd suggest even Larry would draw the line at pointing out such a Grand Replacement Plan was achieving so little!
Think you have the decimal point in the wrong place... surely that should be "0.82% of the target accounts, or 3.3% if they maintain that over the year"
RE: Decimal point
Whoops! Sorry, yes that should have been 0.82% and 3.3%. I can only blame the laughter induced by another TPM "IBM, ra-ra-ra!" article. I suppose we might see a small dent by the time we're all in flying cars. But TPM also didn't go back and read his own article on the IDC figures for the quarter (http://www.theregister.co.uk/2011/05/26/gartner_q1_2011_server_numbers/), which showed that hp grew their UNIX figure by 6.7%. So, if IBM is taking a few accounts, hp seems to be taking a lot of more IBM or Oracle ones!
Itanium is down for the count
"Ignoring the fact that IBM marketting probably counted individual servers"
The number is how many migration factory engagements of moving HP Itanium solutions to IBM.
Most customers move their applications and databases themselves so the number is about 10X higher and does not count the number of customers who stop all new apps on Itanium and are now only buying Power for their databases.
bootnote: Matt....did you notice Intel finally admitted that Kittson will be 22nm? They will still be bringing out Itanic chips that are two fab generations old for the next 6 years assuming they are on time and don't kill the platform
Matt... You kind of read that selective. HP grew their UNIX sales by 6.7% in a marked that rose 20.7%. IBM's UNIX revenue increased by 25.6% and Oracle had a stunning quarter with sales rising 34.4%. It wasn't HP's markedshare that rose with 6.7%.
So HP lost a fair chunk of markedshare, which combined with Oracle's raise in markedshare shows that their foul play towards HP is working. But calculating a bit back then in raw numbers then HP sold for 40 MUSD more in Q1 2011 compared to Q1 2010, Oracle for 156MUSD and IBM for 243 MUSD.
And IBM is pretty close to being as big as HP and Oracle put together in the UNIX marked. They actually might be that looking at Full year 2011.
I don't particular like the fact that Oracle's tactics with drawing support for platforms seem to work.
RE: Selective reading - to a selective conclusion
You are making the unsubstantiated conclusion that any IBM gains were due to competitive take-outs, whereas IBM's gains int eh quarter were more likley to have been old Power upgrades to newer Power6 (which was reduced to clear) and Power7. And probably also because IBM's Power sales took such a beating during the dip in the economy, so they had a bigger hole to climb out of to get back to normal. The point I was making was that IBM's competitive takeouts were not impacting hp's sales, and that Itanium was still growing in sales.
As regards the growth in Snoreacle sales, growing 34.4% of next to nothing is till not much at all.
RE: Itanium is down for the count
".....The number is how many migration factory engagements of moving HP Itanium solutions to IBM......" Seeing as IBM were offering this to us as a free trial, I can suggest that many of those "engagements" were simply single server trials rather than large-scale replacements. You don't have a number of servers involved, and I'm sure if it was at all impressive then IBM would be shouting it out, which makes me conclude that it probably was mainly just toe-dipping by companies keeping their options open. Try again, little troll.
"....Most customers move their applications and databases themselves so the number is about 10X higher ...." That cuts both ways, which means there could also be a large hidden number of migrations off Power. I wouldn't be surprised as AIX seems as clunky as Hell, it's probably a lot easier going AIX to hp-ux, which has a richer toolset and management options, than vice versa.
"....did you notice Intel finally admitted that Kittson will be 22nm?...." Meanwhile, information on the fab sizing for Pee8 is... oh, there isn't any public information for that. In fact, the roadmap for Pee8 is just a placemarker, which isn't surprising as I hear it's still undergoing a complete redesign, whereas Poulsen is taped out. Trolls in glass houses....
It is what it is.
"You are making the unsubstantiated conclusion that any IBM gains were due to competitive take-outs"
No I am not. I have made no such claim. I have no way of knowing that.
"whereas IBM's gains int eh quarter were more likley to have been old Power upgrades to newer Power6 (which was reduced to clear) and Power7. "
Yes I agree, that this has to be a major factor. We are doing massive migrations from POWER4,POWER5 and ..yes also POWER6 to POWER7. And furthermore we are more or less only using POWER 770's as target platform for that. That is one sweet machine when it comes to TCO. Ok I have gotten my first POWER 795 Monster.
But as much of the installations is actually downsizing 690's 595's etc that becomes 770's, which are quite cheap compared to the old machines, and the fact that IBM increased UNIX sales in raw $ numbers with more than the rest of the vendors put together. They they have to have been shipping a sh*tload of tin. And just replacing existing machines is IMHO not the whole story.
So to quote a famous coach. "It is what it is".
Nobody would believe him
IBM has a reputation for never abandoning their customers. Even on a desktop OS like OS/2, they passed that test. They actually paid to Netscape, Adobe to provide software free to OS/2 users and later, they helped Mozilla big time to support OS/2 for releases.
Oracle can't claim to anyone who compiled single software let alone wrote a single line "Hello World" that they aren't maliciously dropping Itanium support.
oracle to DB2
Oracle to DB2 migration is nightmare, and for any mission critical app (and we're talking about this kind of apps here) it means project for many months and damn expensive consultants working 24x7 for weeks to make it work properly. I don't believe they will succeed in selling such a thing in any reasonable amounts. They can sell some power systems, with oracle, to replace old HP systems, true, as long as Larry will keep supporting power. But I guess people will rather prefer to save some money and go for xeons instead. But then, it's HP who has definitely best x86 offer at the moment. And it's as close to business critical as it never been before. I don't think HP will be the big looser here. Just changing the pocket to which money is flowing.
easy for SAP
just ask Pepsi they moved to DB2/Power for SAP from Oracle/SAP/Superdome..... they told Larry to piss off
I wasn't talking about SAP. SAP was always relatively easy to be moved between different supported platforms, I was talking about large, critical, proprietary applications.
Why don't you have info on IBM's plans?
I have seen these claims about IBM's plans not being public. At worst, they are narrowcast under NDA, but there is NOTHING on the Itanium post-Kittson plans. THOUSANDS of people know IBM's roadmap, but none of them write for this website. Remarkable!
IBM is doing Power 7+ this fall. They will boost the number of cores per chip (from 8 to at-least 12), GHz go up at-least 15% (from 3.3,3.6,4.1 to ~3.8,~4.2,~4.7) and die-shrink to 32nm. Power 8 is due in 1H2013 with another die-shrink, more-cores and jacking-the-GHz. Power 8+ is ~ 18 months after that.
Further recessions could defer some activities, but IBM has been close to this schedule for 6 years (Power 7 in 1H2010, Power 6+ in 2H2008, Power 6 in 1H2007, Power 5+ in 2H2005).
RE: Why don't you have info on IBM's plans?
You are asking the wrong question. Intel are standing up and publicly saying "We will make this." What you should be asking is why is Intel ready to make its plans public but IBM are not? The answer is because IBM doesn't want to be held to those plans when they slip or get changed or even cancelled. Go look at the slideset the IBM salesgrunt showed you, check for a footer phrase similar to "Subject to change" or the like. IBM will indemnify themselves at any of these roadmap sessions by telling you, as part of the NDA, that IBM will not be liable for any losses should you make business plans contingent on the roadmap actually coming to reality. In short, any current secret IBM roadmap session has no certainty to it whatsoever. Do you really think Big Blue fanbois like TPM haven't already seen the IBM NDA Power roadmap? Of course they have, they just know it's as vague as the Sunshine one for Oracle's CMT line.
All vendors slip their dates...
Intel has certainly 'stood up and announced' many dates and subsequently recast those predictions. Often with minimal apology. The is Intel's culture. I guess that you favor it.
IBM doesn't do this as publicly and it is pesky. I agree with that.
But to write that IBM doesn't send out its plans is taking it to another level...
- Geek's Guide to Britain INSIDE GCHQ: Welcome to Cheltenham's cottage industry
- 'Catastrophic failure' of 3D-printed gun in Oz Police test
- Game Theory Is the next-gen console war already One?
- Analysis Spam and the Byzantine Empire: How Bitcoin tech REALLY works
- Apple cored: Samsung sells 10 million Galaxy S4 in a month