There’s nothing like a vendor war to keep things interesting, and Oracle has launched a great one with its recent decision to stop development for future Itanium processors. According to Oracle, Itanium is the deadest of dead parrots - not pining for the fjords, not tired from a prolonged squawk, but well and truly dead. Intel …
More Questions No Answers!!!!!!!!!!!!!!
Now it is sounding like Oracle has a point. If there is convergence of the technologies in Xeon and Itanium and as you put it "the choice between Xeon and Itanium processors in the future will be dictated by customer O/S choice. If you’re running HP-UX or NonStop, you’ll run Itanium processors; for Windows or Linux, you’ll buy Xeons.", why not force HP to port HP-UX or NonStop to run on XEON. Why should Oracle continue to develop for both platforms.
I am no expert on this issue and is neutral, but I have never heard any serious positive comments about Itanium over the years. In fact many expected it would have died already. But since Oracle's announcement I have heard only Oracle bashing without any serious discussions on the technical side of their argument.
So my request here is to be educated. Why should Oracle continue to invest in Itanium a possible dying platform if it is converging with Xeon. If it is as some suggested, HP Integrity Servers are predominantly purchased to run Oracle DB, why should Oracle not force the market to move to Xeon based system if they are expected to at worst offer comparable performance in the near future. oR if you have to run Oracle DB on a nix why not force the market hand to implement Solaris on comparable hardware instead.
What if any are the advantages of HP Integrity Servers if Xeon based system in the future will offer comparable features.
I have more questions and no answers on this one and am begging for technical discussion rather then the flames that characterize the posts of the past weeks.
its about volume
Perhaps SPARC and Itanium will limp on for several more years in legacy shops but the writing is on the wall. I would say the same thing about POWER if not for IBM's stubbornness in remaining a chipmaker as well as the volumes they do get on production for game consoles. The fact is simply the benefits of high end architectures against x86_64 are not worth the premium you have to pay (especially with more and more RAS features coming to x86 each generation). Now ARM on the low end may end up be a different story.
only problem SPARC blows
>Or is this the first move in a plan to use its power as a dominant enterprise ISV to move customers to an all-Oracle hardware, o/s, and software stack?
Even with a market leading position in the database sphere only releasing their software on the garbage architecture that is SPARC would be suicidal. They might cut off POWER but no way can they afford to cut off x86 (the majority of their market) and thus many of their competitors as well. The hardware play to increase margin maybe looks good on paper but the fact of the matter is they did inherit a dead horse architecture in SPARC.
Larry may live to regret some of his recent decisions
Why do I have the gut feeling that Larry may soon have to regret his decision to make Sun x86 servers an Intel only shop. Is it something in the air?
Some objectivity please?
Dan, as representing Gabriel Consulting surely you need to be a bit objective?
This is your 2nd article on this theme and it would perhaps be pertinent to highlight precedent and other relevant facts:
Itanium: remember the Itanium Alliance? Since 2005 they have lost 5 of the main 6 players (Hitachi, Fujitsu, NEC, Unisys, Bull and Silicon Graphics). That is a horrific attrition rate. HP is last man standing (and the word on the street is their sales teams are moving customers frantically to x86 as Oracle and IBM plunder the user base left hanging because HP has shrunk the portfolio drastically including the move to the weird blades with their front mounted interconnects)
Intel: The public roadmap for Intel of 24 slides gives 1(and 1/4!) slides to Itanium. Kitson is one word, no dates, not even year or decade, no details, nadah, zero, none at all... (Google for the corresponding SPARC roadmap at oracle.com and it shows 3 generations out with a decent summary of major features for each. Plus there are webinars where execs talk through approximate dates)
HP: HP's 2 closest partners for OSs and (former?) members of the rapidly dwindling Itanium Alliance, Microsoft and Redhat both abandoned Itanium long before Oracle and that doesn't warrant a mention? It gives both context and precedence.
Maybe a little more balance and background detail?
So Oracle is following a trail well blazed by both Itaniums hardware & software partners. You should at least mention that?
(Oh and for the record all of the next gen games consoles have abanded the Power instruction set and they *never* used a P6 or a P7 anyway so your are drinking the kool aid. Go over to IDC etc and you'll see IBM UNiX revenue leader, SPARC volume leader. In fact SPARC outsold the entire Itanium alliance even when they where at the peak in CPU volumes. But you know all this you work for Gabriel?)
Whether Dan is "objective" or not, I don't know, but presenting a bunch of FUD as a response certainly isnt...
>> Itanium: remember the Itanium Alliance?
Well all these companies largely depended on Windows/Linux on Itanium, so not surprising they've fallen by the wayside - to be honest Windows/Linux on Itanium was always a side-show, though I'm sure Intel would have preferred it wasn't. Incidentally both SGI and NEC still sell Itanium systems...
>> (and the word on the street is their sales teams are moving customers frantically to x86...
Ah, so Dan should be objective, but you can quote scuttlebut? Care to give some examples of this? This sounds like "wishful thinking" to me, and certainly isn't born out by what I see in terms of industry market share stats or what actions I see from HP salespeople. But that's my _subjective_ opinion (wow see what I did there?)
>> Intel: The public roadmap for Intel ...
Not uncommon on public processor roadmaps - incidentally Power is the same - one word "Power 8". The SPARC roadmap has a lot of interesting figures on it but in my _subjective_ opinion history tells us to be cautious in trusting SPARC roadmaps (Rock) and there's lots of ways of interpreting the 2x/4x/10x numbers given in the public slides.
>> Microsoft and Redhat both abandoned Itanium...
Both OS vendors, not application vendors - there was little money in selling Windows/RHEL on Itanium - But Oracle products on HP-UX? given current market share stats from IDC and Gartner, you would think that Oracle makes as much money out of that as it does out of selling Oracle products on Solaris. It's also interesting that Oracle claim "Oracle is the last of the major software companies to stop development on Itanium" - which is odd, as you would think they would have heard of those tiny little software companies called IBM and SAP?
>> In fact SPARC outsold the entire Itanium alliance even when they where at the peak in CPU volumes.
Hmmm far from telling Dan to go review IDC/Gartner numbers, I suggest you do the same yourself - whilst the whole UNIX server market has shrunk in the past 2 years, SPARC has fallen off a cliff. According to Gartner in Q4 2010 SPARC revenue for dropped by 16.2% and SPARC shipment figures fell by 40%.
Where on earth did you get the idea that Oracle are plundering the HP-UX user base? HP and IBM trade blows continually, but for both vendors vs Oracle it has been only one-way traffic lately. Neither HP nor IBM lose to Oracle these days unless Oracle is already the incumbent! Fact! And both IBM and HP have been picking up disgruntled Oracle customers at an alarming (for Larry) pace. Hence this announcement. The UNIX marketplace is essentially a 3-way tie so there is no market driven reason to cease development of Oracle on Itanium. This is purely an attempt by Oracle to force customers to look at their hardware because customers aren't doing that out of choice these days!
Compaq-inherited OSes: NSK and ... what else?
"If you’re running HP-UX or NonStop, you’ll run Itanium processors; for Windows or Linux, you’ll buy Xeons."
Didn't HP inherit another OS from Compaq in addition to Tandem's NonStop stuff?
Hey Skaugen (and maybe Dan too), here's a hint for you: MVS (some re-assembly required).
The other OS...
"Didn't HP inherit another OS from Compaq in addition to Tandem's NonStop stuff? Hey Skaugen (and maybe Dan too), here's a hint for you: MVS (some re-assembly required)."
I think you are mixed up. MVS is IBM's mainframe OS. The "other Compaq OS" is Tru64 (aka Digital Unix, aka OSF/1). Tru64 is basically dead, as it only runs Alpha CPUs. HP is ending support for Tru64 end of next year. Initially, HP actually planned to merge HP/UX features into Tru64, but and kill HPUX, but they did the opposite.
Besides Intel's lukewarm promotion of Itanium, HP hasn't exactly been creating a lot of excitement around HP-UX either. Support for a maximum of 128 cores sounds pretty rural in 2011.
He's not mixed up - the clue is in there "here's a hint for you: MVS (some re-assembly required).
" - he means VMS aka OpenVMS.
>> HP actually planned to merge HP/UX features into Tru64, but and kill HPUX, but they did the opposite.
No that's incorrect - HP said pretty much immediately when the Compaq merger went through that they would kill Tru64 and move features into HP-UX. They never said they would kill HP-UX in favour of Tru64.
Enterprise UNIX customers don't want "excitement" from their OS - they want a stable/mature platform - it may be "boring" but that's what they want...
MVS? SMV? Oh, what was that OS called that WNT was meant to grow up to?
Yes Tom, I know MVS isn't and wasn't an HP OS. Hence "some re-assembly required".
When you reassemble MVS slightly differently enough times, you eventually get:
The other VAX, Alpha, Itanium (and x86-64) OS, the OS that Skaugen forgot to mention.
The other VAX, Alpha, Itanium (and x86-64) OS
Silly me, failed again, I forgot a questionmark, that should obviously have read:
"The other VAX, Alpha, Itanium (and x86-64?) OS"
Your calling is not in journalism
Maybe you would be better suited working for a PR firm rather than posing as a journalist...you only embarass the rest of us.
- Geek's Guide to Britain Kingston's aviation empire: From industry firsts to Airfix heroes
- Analysis Happy 2nd birthday, Windows 8 and Surface: Anatomy of a disaster
- Adobe spies on readers: EVERY DRM page turn leaked to base over SSL
- Google chief Larry Page gives Sundar Pichai keys to the kingdom
- Breaking news: Google exec veep in terrifying SKY PLUNGE DRAMA