backtop


Print 29 comment(s) - last by ekv.. on Mar 26 at 5:20 AM


Oracle says no to Itanium  (Source: Intel)
Intel and HP are not happy

Intel has more processors than most consumers are aware of spanning categories that have nothing to do with notebooks, netbooks, and desktop computers that we use at home and in the office. Intel has a full line of Xeon and Itanium processors that it offers to computer makers for use in servers for businesses.

The Xeon processor line is Intel’s volume server processor line, while the Itanium processor --which isn't based on the x86 architecture -- sells in much lower volume for niche use in high-end data processing systems for financial, medical, and other uses. The Itanium processors are based on EPIC architecture, and do not support x86 software and operating systems.

That lack of support means that the companies that make software like Microsoft, Red Hat and Oracle have to develop versions of their software specifically to run on the Itanium processors. Oracle announced this week that it will stop developing new software for Itanium-based processors. Red Hat and Microsoft have previously made the same move to stop support. EWeek reports that Oracle decided to stop Itanium development after talks with people within Intel indicated that the chip giant was looking to wind down Itanium production.

A statement from Oracle said, "Intel management made it clear that their strategic focus is on their x86 microprocessor and that Itanium was nearing the end of its life."

However, Intel CEO Paul Otellini strenuously denies that statement. Otellini says that Intel is working hard on developing new Itanium processors and that processors are in the pipe and on schedule in the Itanium family.  The next generation Itanium processor is called Poulson and will reportedly use a new architecture that will allow Itanium to continue for years to come.

Otellini said, "Intel's work on Intel Itanium processors and platforms continues unabated with multiple generations of chips currently in development and on schedule. We remain firmly committed to delivering a competitive, multigenerational roadmap for HP-UX and other operating system customers that run the Itanium architecture."

EWeek reports that Poulson will be a 32nm chip with up to eight cores with twice the performance of the Tukwila Itanium chip in use now. The follow up to Poulson called Kittson is also under development according to eWeek. HP is the largest user of Itanium processors and is not happy with Oracle's decision either.

HP's David Donatelli said, "We are shocked that Oracle would put enterprises and governments at risk while costing them hundreds of millions of dollars in lost productivity in a shameless gambit to limit fair competition."



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Itanium is doing well actually
By finalfan on 3/24/2011 1:22:03 PM , Rating: 2
From http://en.wikipedia.org/wiki/Itanium#Market_share

For the combined POWER/SPARC/Itanium systems market, IDC reports that POWER captured 42% of revenue and SPARC captured 32%, while Itanium-based system revenue reached 26% in the second quarter of 2008

I don't see there is any problem for Itanium in its own market segment.

In Oracle's theory, they should stop writing software for SPARC as well.




RE: Itanium is doing well actually
By fic2 on 3/24/2011 1:35:40 PM , Rating: 5
Except that:
A) Oracle now owns SPARC
B) Oracle probably makes more money on SPARC because of A
C) Oracle has a say in the SPARC future because of A
D) How do you know the break even point isn't 32%?
E) AFAIK Sun (now part of Oracle) never made Itanium systems
F) System revenue has nothing to do with software revenue
G) You report is 3 years old


RE: Itanium is doing well actually
By finalfan on 3/24/2011 1:44:05 PM , Rating: 2
Your points are valid.

But what I want to point out is Oracle is not telling all the truth. It's not Itanium is dying. It's Oracle doesn't want to support competition's platform. Although there is nothing wrong about the move. Only its statement made itself ugly, as always.


RE: Itanium is doing well actually
By fic2 on 3/24/2011 1:50:48 PM , Rating: 3
According to wiki (http://en.wikipedia.org/wiki/Itanium) Intel has stopped support for Itanium in it's Fortran and C/C++ compilers. If Intel won't even support it's own chip why should Oracle.

I doubt that Oracle will stop supporting x86 which is a competitors platform, too. It looks like they also support the IBM Power platform.


RE: Itanium is doing well actually
By finalfan on 3/24/2011 2:11:49 PM , Rating: 2
Every new major release of Intel compiler is to support new processor and intruction set. Since there is no new Itanium released in that period, there is no reason to release a new version for it. The previous release 11.1 is still available for purchase. And Intel is a major contributor to GCC. Of course, that's just my opinion.

Since HP is almost of sole system vendor of Itanium and HP is taking more serious in Software with previous SAP CEO in the helm, and also because of you-know-what-happened between Oracle and HP, the move is not a big surprise at all.


RE: Itanium is doing well actually
By fic2 on 3/24/2011 3:45:03 PM , Rating: 2
Oracle has said that it will continue to support existing Itanium customers, too.

BTW, Intel has also canceled their yearly EPIC workshop:
http://www.cgo.org/cgo2011/epic9/

Intel is saying one thing but seems to be doing the opposite.


RE: Itanium is doing well actually
By finalfan on 3/24/2011 4:55:50 PM , Rating: 2
There is no EPIC workshop in 2003 and 2009 either. So...


RE: Itanium is doing well actually
By JediJeb on 3/25/2011 1:42:27 PM , Rating: 2
Another thought, didn't the previous president of HP take a job with Oracle? Makes you wonder that Hurd might have to do with all this. If you want to bet back at your former employer just do something to help kill off an entire line of their products.

Let the conspiracy theories begin ;)


RE: Itanium is doing well actually
By ekv on 3/26/2011 5:20:40 AM , Rating: 2
Nice jab!


Yep
By dagamer34 on 3/24/2011 12:51:57 PM , Rating: 2
If it wasn't dead before, it certainly is now.




RE: Yep
By Motoman on 3/24/2011 1:13:49 PM , Rating: 4
Oh, it's been dead for years...it just didn't know it yet.


RE: Yep
By Phynaz on 3/24/2011 1:29:56 PM , Rating: 1
And yet it's sales have been increasing every year.

Hmmmmm


RE: Yep
By fic2 on 3/24/2011 1:36:16 PM , Rating: 3
Source?


RE: Yep
By GodisanAtheist on 3/24/2011 3:21:57 PM , Rating: 2
The living dead, then.


RE: Yep
By spread on 3/24/2011 1:47:19 PM , Rating: 3
quote:
Oh, it's been dead for years...it just didn't know it yet.


Intel Itanium: Directed by M. Night Shyamalan


RE: Yep
By fic2 on 3/24/2011 6:22:41 PM , Rating: 1
The Dead Collector: Bring out yer dead.
[a man puts a body on the cart]
Large Man with Dead Body: Here's one.
The Dead Collector: That'll be ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: What?
Large Man with Dead Body: Nothing. There's your ninepence.
The Dead Body That Claims It Isn't: I'm not dead.
The Dead Collector: 'Ere, he says he's not dead.
Large Man with Dead Body: Yes he is.
The Dead Body That Claims It Isn't: I'm not.
The Dead Collector: He isn't.
Large Man with Dead Body: Well, he will be soon, he's very ill.
The Dead Body That Claims It Isn't: I'm getting better.
Large Man with Dead Body: No you're not, you'll be stone dead in a moment.
The Dead Collector: Well, I can't take him like that. It's against regulations.
The Dead Body That Claims It Isn't: I don't want to go on the cart.
Large Man with Dead Body: Oh, don't be such a baby.
The Dead Collector: I can't take him.
The Dead Body That Claims It Isn't: I feel fine.
Large Man with Dead Body: Oh, do me a favor.
The Dead Collector: I can't.
Large Man with Dead Body: Well, can you hang around for a couple of minutes? He won't be long.
The Dead Collector: I promised I'd be at the Robinsons'. They've lost nine today.
Large Man with Dead Body: Well, when's your next round?
The Dead Collector: Thursday.
The Dead Body That Claims It Isn't: I think I'll go for a walk.
Large Man with Dead Body: You're not fooling anyone, you know. Isn't there anything you could do?
The Dead Body That Claims It Isn't: I feel happy. I feel happy.
[the Dead Collector glances up and down the street furtively, then silences the Body with his a whack of his club]
Large Man with Dead Body: Ah, thank you very much.
The Dead Collector: Not at all. See you on Thursday.
Large Man with Dead Body: Right.


RE: Yep
By tamalero on 3/25/2011 11:53:49 AM , Rating: 2
WHAT A TWEEEST!


65 to 32 nm and only 2x speedup?
By DanNeely on 3/24/2011 1:25:28 PM , Rating: 2
quote:
EWeek reports that Poulson will be a 32nm chip with up to eight cores with twice the performance of the Tukwila Itanium chip in use now. The follow up to Poulson called Kittson is also under development according to eWeek. HP is the largest user of Itanium processors and is not happy with Oracle's decision either.


Tukwila is a 65nm design. Going to 32nm drops transistor size by 75% (area), which ignoring architecture improvements and faster clock speeds should allow a 4x speedup from more cores. Is Poulson just a significantly smaller chip (Tukwila was huge) or is this an indication that Intel's having major problems with it?




RE: 65 to 32 nm and only 2x speedup?
By finalfan on 3/24/2011 1:36:18 PM , Rating: 2
Tukwila is four core and Poulson is 8 core so there is 2x speedup.

There is no direct link between performance and node process.

See http://en.wikipedia.org/wiki/Itanium#Poulson

And keep in mind Tukwila is a 12 wide-issue processor. The Core 2 series is only 4 issue. IPC wise, it's much much faster than the everyday commodity CPUs.


RE: 65 to 32 nm and only 2x speedup?
By DanNeely on 3/24/2011 4:24:58 PM , Rating: 2
The die shrink alone has room for 4x the transistors though. If Poulson cores are no larger than tukwila that means the die size is 50% smaller (possible because of how big Tukwila was), or if the die size remained the same that despite being twice as large there wasn't any speedup in performance at all.

The fact that they're only claiming 2x implies that the new architecture is no faster than the old one, which seems surprising unless something went wrong. If nothing else the increase in the width should have given a performance boost as well.


By finalfan on 3/24/2011 5:22:44 PM , Rating: 2
The information for Poulson is rather scarce. The Poulson die is smaller than Tukwila (544 vs 698) but with 1.5 times of transisters (3.1B vs 2B). There is no mention of the clock speed though. So it's hard to deduct any performance number from there. The 2X speedup is mostly a guess and assumed the minimum. Since Intel didn't release any data for it, we have to wait till it's available. We will see how much improvemnet from 6 issue to 12 issue can speed up the single thread performance.
The point is if Intel is to abandon the platform, there is no way they could've presented the new architecture in ISSCC just last month. Actually nobody was expecting them to do so.


Hurds revenge ...
By haplo602 on 3/25/2011 10:56:24 AM , Rating: 3
Oracle is simply trying to increase it's own income. What do you think will be suggested as an Itanium replacement architecture ? Sparc of course.

And since HP is the largest (and almost exclusive) Itanium HW vendor, this can be seen as Hurds revenge.

Anyway the x86 world can only dream about RAS capabilities of Itanium based hardware.




RE: Hurds revenge ...
By JediJeb on 3/25/2011 1:47:44 PM , Rating: 2
I was wondering the same thing. Maybe he is trying to knock out a whole line of HPs products in one blow.


WebLogic
By Slaimus on 3/24/2011 1:49:40 PM , Rating: 3
This is probably another case of Oracle dumping "niche" markets from their acquisitions.

Prior to Oracle's purchase, WebLogic was the strongest IA64 supporter among the Java EE vendors. They had IA64 tuned versions of their JRocket JDK that were used in most of top scoring Itanium benchmark scores.

This is similar to Oracle stopping Opteron servers after Sun's purchase, even though Sun was the most vocal supporter of AMD64 servers for most of their existence.




RIP Itanic
By Pirks on 3/24/2011 12:46:49 PM , Rating: 2
Rest In Pain ;)




RIP Itanic
By dsx724 on 3/24/2011 12:47:57 PM , Rating: 2
You sucked from the very start and you never learned the lesson taught by your forefathers about compatibility.

http://en.wikipedia.org/wiki/File:Itanium_Sales_Fo...




Oracle sales on Itanium?
By nafhan on 3/24/2011 1:32:44 PM , Rating: 2
So, the real question for me... how much Itanium software was Oracle actually moving? I get the feeling that it might be really small volume even relative to the low volume Itanium software industry. This makes me feel that Oracle is trying to hide their own failures behind the failures of others.

At the same time, it does seem like Intel needs to go ahead and let the Itanic sink at some point. I think they're making back the money spent on manufacturing, but that's about it.




By errolDC on 3/24/2011 10:08:56 PM , Rating: 2
Itanium's been dead for years.. We have just been waiting for the body to hit the floor.




What?
By fic2 on 3/24/2011 1:29:28 PM , Rating: 1
quote:
HP's David Donatelli said, "We are shocked that Oracle would put enterprises and governments at risk while costing them hundreds of millions of dollars in lost productivity in a shameless gambit to limit fair competition."


This guy is an idiot. How is Oracle not producing something "limit[ing] fair competition"?

If gov'ts and enterprises put their money on a horse that has been dead for years that is their fault not Oracles.




"The Space Elevator will be built about 50 years after everyone stops laughing" -- Sir Arthur C. Clarke











botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki