Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

HP Asks Judge To Enforce Itanium Contract Vs. Oracle

Soulskill posted more than 2 years ago | from the oracle-making-a-play-for-more-courtroom-marketshare dept.

HP 124

Dupple writes with this quote from a Reuters report: "Hewlett-Packard Co told a judge on Tuesday that Oracle Corp should be ordered to make its software available on HP's Itanium-based servers for as long as HP sells them. Lawyers for HP and Oracle presented closing arguments in a California state court for the first phase in a bitter lawsuit between the two tech giants. ... Oracle decided to stop developing software for use with Itanium last year, saying Intel made it clear that the chip was nearing the end of its life and was shifting its focus to its x86 microprocessor. But HP said it had an agreement with Oracle that support for Itanium would continue, without which the equipment using the chip would become obsolete. HP said that commitment was affirmed when it settled a lawsuit over Oracle's hiring of ousted HP chief executive Mark Hurd. HP seeks up to $4 billion in damages."

Sorry! There are no comments related to the filter you selected.

Whoa (-1)

Anonymous Coward | more than 2 years ago | (#40472299)

Itanium still exists? Why would anyone admit to that?

Re:Whoa (1)

vlm (69642) | more than 2 years ago | (#40472359)

Apparently its being sustained by the booming GCOS 8 market. Those two customers must be very happy.

Indeed (0)

Anonymous Coward | more than 2 years ago | (#40478153)

These two customers haven't been pwned by Chinese Intel, different to everybody else.

SOS (4, Funny)

plover (150551) | more than 2 years ago | (#40472341)

The Itanic is sinking!

Re:SOS (1)

Sponge Bath (413667) | more than 2 years ago | (#40473115)

Near, far, wherever you are, I believe that the heart does go on... at least until Oracle snuffs it.

The ghost of Ken Olsen called, (4, Insightful)

soupforare (542403) | more than 2 years ago | (#40472387)

he extends sympathies.

damages per processor (3, Informative)

vlm (69642) | more than 2 years ago | (#40472389)

HP seeks up to $4 billion in damages.

Years ago when the itanic was sinking, I heard shipping estimates as low as 200K processors annually. I'm sure its lower now. But that implies something on the order of $20K damages per processor shipped, which is astounding.

Re:damages per processor (5, Interesting)

Anonymous Coward | more than 2 years ago | (#40472473)

Where I work, we use Alpha, and have investigated moving to Itanium, but with this debacle, we won't make the move, as we use also Oracle. I won't discuss actual numbers, but $20,000 per CPU is in the ballpark of our annual support costs (if you include hardware and software). At some point we will either port or rewrite, and in either case, it won't be an HP platform we land on, but we may still use Oracle for the database. This case has very real implications for HP, and hurts them more than it hurts Oracle.

Re:damages per processor (1)

Anonymous Coward | more than 2 years ago | (#40472927)

I would be interested in knowing what OS you have been using, VMS or Digital Unix maybe? You probably will have much bigger issues than just which new hardware platform you will be using. DEC is gone. SGI is gone. HP is slipping away. Sun Micro was rescued. Oracle might just be the best source for HW (Sun), OS (Win64, Solaris), and DB (Oracle DBS) combined.

Oh, and I agree with you about the $20K / CPU, although that might be a bit on the low side for support costs (volume discount?).
HP sacrificed their 64bit processor in favor of the Itanium (that's what SGI did, too). Who still fabricates the Alpha?

Re:damages per processor (1)

Anonymous Coward | more than 2 years ago | (#40473311)

Alpha was available new for longer than you might think, at least for current HP customers. I believe 2006 was the cutoff for new alpha hardware. And it is still supported, through I believe at least 2014. and probably longer. We use VMS, and HP is still regularly patching an updating the OS. That is more than I can say for RDB, which started out as a DEC product and ended up in the hands of Oracle. We still use RDB, from back in the DEC days. But there are plenty of people running the numbered Oracle database products on VMS as well. In our case, we are slowly migrating most applications to completely different platforms. I'm sure it will be difficult and time-consuming, but there are lots of hardware spares, and VMS is a wonderfully fault-tolerant operating system. The clustering is still second to none. It is probably my favorite server OS, from a support perspective. This is not to say I enjoy dealing with HP's support; on the contrary it's a miserable experience, and getting more miserable with each passing quarter.

The MBAers destroyed HP (0)

Anonymous Coward | more than 2 years ago | (#40478197)

..but it was David Packard's fault to believe into the "good managers can manage everything" meme. He should never have made Lew Platt CEO. This man outsourced the CPU R&D to Intel and compiler R&D to some random Indians, IIRC.

Re:damages per processor (0)

unixisc (2429386) | more than 2 years ago | (#40478467)

Alpha's demise happened while it was still w/ Compaq. After NT was discontinued on the Alpha, that platform didn't stay long, since Compaq decided to go along w/ the Itanic as well. AFAIK, it was under Compaq that the migration started of OVMS from Alpha, and Tandem NonStop from MIPS, both to Itanic. So the GGP may have been using OVMS/AXP and may have investigated going to OVMS/Itanic.

It doesn't make sense to go from one legacy platform to another whose future was never really there. Itanium would have been a fine platform for new native software, under native OSs like HP/UX, Debian Linux and FreeBSD. It's horrible for supporting legacy platforms. Right now, it's no different as an industry standard from what PA-RISC was - a single sourced HP platform. Just about everybody who had adapted it - SGI, Dell, et al have abandoned it. In fact, even HP discontinued workstations based on this years ago. Only reason to go w/ Itanium is if one is starting from scratch, and is going to implement FOSS across the board, so that they're not left @ the mercy of suppliers like Oracle. Get an HP Integrity server, load it up w/ FreeBSD and then install whatever open source software they need and can get their hands on.

If the GGP's h/w is still good, they could run it as long as possible, but going forward, it's good to migrate to a different platform, w/ Open Source software, so that they won't be left high & dry next time. One option could be SparcServers, but use something like MySQL instead of Oracle. Other option - go for IBM. But here too, go w/ open source software as much as possible, so that if IBM ever discontinues support, the customer would still be able to get expertize to keep it running as long as possible.

Re:damages per processor (1)

Billly Gates (198444) | more than 2 years ago | (#40473837)

It is amazing Oracle is still in business. If it were not for the billions of lines of code specifically for crapware enterprise software it would be dead. Was it always this much or did Larry jerk up the prices this high after customers were hooked?

To me Oracle DBMS is the IE 6 of databases. It is out of date, its sql is proprietary, has proprietary tools, and the only reason people use it like IE 6 is because code depends on it and these saps can't leave.

Maybe when NoSQL matures it will eat it for lunch. There are places that use PIC still which is supperior to SQL that has been popular for 40 years.

Re:damages per processor (1)

nitehawk214 (222219) | more than 2 years ago | (#40474627)

I wouldn't pin the insanely large amount of crap enterprise software being tied so tight to Oracle as being solely Oracle's fault. Bloated enterprise software will always find terrible platforms to become tied to. Companies like Oracle, IBM, Microsoft live off of vendor lock in and encourage it; but the managers at large software projects seem to want lock in as well.

I could never really understand this effect, even large organizations that moan about how horrible it is to be locked in to Microsoft and Oracle willfully dive right in to it even when starting new projects. Fortunately I now work at a small company with management direction on the best product for the money, including looking forward at being gouged by support contracts long term.

Re:damages per processor (1)

Billly Gates (198444) | more than 2 years ago | (#40474753)

It is funny they want lockin until companies like MS stop XP support or when Oracle keeps upping its requirements and costs. Then its holy shit how did this happen!!

Free software like FOSS offers alternatives and noSQL is new and open alternatives like PostgreSQL and mysql greatly reduce this. I do no understand people who think average Joes should run Linux on the desktop but for the server this makes sense.

No lock in and vendors and users encourage compatibility. The fact that browser vendors finally agreed to W3C standards means we can finally bury nightmares like IE 6 behind. I think managers afraid of change want it frozen only at that moment when it is new. Good managers want standards so they are not tied later on

So you punish HP (0)

Anonymous Coward | more than 2 years ago | (#40478173)

..because Oracle is a disloyal POS ?? Makes a lot of sense.

damages per processor ... WHY? (4, Insightful)

DragonWriter (970822) | more than 2 years ago | (#40472535)

Years ago when the itanic was sinking, I heard shipping estimates as low as 200K processors annually. I'm sure its lower now. But that implies something on the order of $20K damages per processor shipped, which is astounding.

Why would you even think of damages in terms of "per processor shipped" (and, even worse, in terms of annual processor shipments)? Even assuming the estimates you refer to are accurate, the computation you make is meaningless.

Re:damages per processor (1)

the_humeister (922869) | more than 2 years ago | (#40472563)

Yet Intel's Itanium profits are greater than all of AMD's. Wish AMD would do better financially.

Re:damages per processor (0)

Anonymous Coward | more than 2 years ago | (#40473189)

I wish AMD would make a processor worth buying! Then I'd gobble them up like Skittles!!

Re:damages per processor (0)

Anonymous Coward | more than 2 years ago | (#40475689)

But not iced tea and Skittles, else you might get shot.

Re:damages per processor (1)

Lehk228 (705449) | more than 2 years ago | (#40473323)

$20K/core is what oracle charges right?

Re:damages per processor (3, Funny)

moderatorrater (1095745) | more than 2 years ago | (#40473455)

Maybe for the student version.

Re:damages per processor (1)

Billly Gates (198444) | more than 2 years ago | (#40473807)

The metrics probably include the price of the server as well as lost HP management software it could have sold. Yes the latter part is always inflated but Itanium was never made for desktops but really expensive 8 - 128 cpu servers.

Blades came out later and helped kill the Itanium as they were old fashioned big boxen but still this is what HP probably used to estimate the damages of lost sales.

Re:damages per processor (1)

Fjandr (66656) | more than 2 years ago | (#40476337)

I would guess so, since the most recent numbers I heard put their shipped Itanium systems at something like $20k-$250k, depending on configuration.

Granted, it's been a couple months and I wasn't terribly interested in the article, so I could be off by quite a bit, but HP's made a continued investment because they are actually still making money servicing whatever niche is filled by those Itanium systems.

Re:damages per processor (2, Interesting)

Anonymous Coward | more than 2 years ago | (#40476693)

One of our customer bought some Itanium servers from HP, 3-year old unused hardware, they must have done a deal they couldn't refuse. Then the Oracle wouldn't run fast enough and of course it was our problem, us being the software vendor. After expensive CPU upgrades and everything else, it still runs like a dog. For the fraction of the money, they could have got some intel kit and run Linux with 10x performance.

I doubt it's Itanium's fault here (0)

Anonymous Coward | more than 2 years ago | (#40478259)

Itanium is generally considered to be in the same rough performance "ballpark" as x86. If you can't make it run properly on Itanium+HPUX, it is probably your incompetence. I was once an HP employee (until 1997) and their machines were always competitive. Why do you think they have (historically) been so successfully ? Only because of their superior salesforce ?
I am telling you a secret; their PA-RISC CPUs were performance leaders in the mid-nineties.

Re:damages per processor (1)

unixisc (2429386) | more than 2 years ago | (#40478575)

Uh, Debian Linux could have been run on those HP Itanium servers as well. Or one could have used the last version of Centos that worked on the Itanium. Where did people get this idea that Linux is only good for Xeon or Opteron platforms?

Currently, the Itanium supports

  • HP/UX - HP's own Unix back from the PA-RISC days
  • Debian Linux
  • FreeBSD

So just b'cos someone buys a de-facto proprietary hardware from HP doesn't mean that they are locked on to HP/UX and whatever is in their mercy. They could have installed, say, FreeBSD and then ProgreSQL and any other FOSS that they needed, and started w/ that. That too would have given them 10x performance w/o going through CPU upgrades and other expenses.

Why exactly ? (1)

Superdad (847315) | more than 2 years ago | (#40472393)

Can somone opine as to why exactly HP is doing this ? What do they hope to get ? Why dont they simply cut n run, and/or move HP-UX to x86 as they've already proven to themselves that it can run on x86.

Corporations are weird.

Existing Customers (4, Insightful)

raehl (609729) | more than 2 years ago | (#40472485)

HP sold Itanium boxes to customers who use them to run Oracle. Oracle stops supporting Itanium and the customers are stuck holding computers that don't do what they paid for them to do.

There's probably penalties in HPs contracts with their customers in the event of such a circumstance. Or maybe they just don't want their customers to feel like HP screwed them.

Re:Existing Customers (2)

AmiMoJo (196126) | more than 2 years ago | (#40472585)

Even if HP forces Oracle to continue software development how much effort will they really put into it? If some critical bug only affects Itanium how many resources will be thrown at it?

Re:Existing Customers (1)

Aighearach (97333) | more than 2 years ago | (#40473295)

Even if HP forces Oracle to continue software development how much effort will they really put into it? If some critical bug only affects Itanium how many resources will be thrown at it?

In that case if it causes HP additional damages, they'll just have to pay whatever they're costing HP. They'll have already lost.

Re:Existing Customers (5, Interesting)

Bert64 (520050) | more than 2 years ago | (#40472843)

HP have been screwing their customers for a while already...
Killing PA-RISC and Alpha, forcing customers onto IA64 which now looks like it will die soon too...
With each processor transition, customers at the very least have to recompile their code or run it under slow emulation...

Sun and IBM snapped up a lot of customers from HP over the IA64 transition, and i fully expect them to do the same when HP finally abandons IA64 and moves its customers onto something else.

Introducing a new, incompatible CPU was never going to work...

Too much code is only available in binary form, and so ran on ia64 very slowly under emulation. Vendors had no incentive to port their code because there were so few users, and users had no incentive to buy into the architecture because there was no software. Who wants to pay 5 times more for a cpu that when saddled with emulation is actually slower than the cheaper cpus?

The hardware was never cheap enough to attract hobbyist developers, so even open source code was often not well tested on them (nor are new risc cpus from other vendors, but old ones can be had cheaply). Had the hardware been cheap, it would have gained a lot more traction in the linux market at least.

Re:Existing Customers (1)

laffer1 (701823) | more than 2 years ago | (#40473641)

You are absolutely right. I looked into purchasing an older IA64 box on ebay last year. Prices were still to high to justify the cost for something I barely know anything about. Then trying to figure out what OS to run on it was another problem. I would have ended up with FreeBSD, but it's certainly not a viable platform at this point.

Are You $hilling ?? (0)

Anonymous Coward | more than 2 years ago | (#40478317)

I doubt any customer would have moved from HP to the snail-fast SPARC abominations. Maybe to IBM, that sounds plausible.
Also, for most software vendors, Itanium support was just a matter of rebuilding their app on the new hardware. I remember running all sorts of GNU tools on HPUX / PA RISC and all of it worked just nice. Gcc to ghostview.

Re:Existing Customers (2)

Kozar_The_Malignant (738483) | more than 2 years ago | (#40473357)

>Or maybe they just don't want their customers to feel like HP screwed them.

Where, in the last three years, can you find a customer who doesn't feel screwed by HP?

Re:Existing Customers (1)

nitehawk214 (222219) | more than 2 years ago | (#40474645)

... Or maybe they just don't want their customers to feel like HP screwed them.

They both HP hardware, what did they expect?

I am not familiar with high end licensing schemes, but I have a feeling HP would make more money on support contracts for these behemoths than selling them new hardware. Failing that, I am sure a pretty good percentage of customers getting screwed on Itanium would take their support contract money to other vendors when it comes time to buy new hardware.

Re:Why exactly ? (4, Interesting)

Sique (173459) | more than 2 years ago | (#40472501)

Because they have contracts with their customers guaranteering them continued software support. And if the main supplier stops software support, those contracts become quite shaky.

Re:Why exactly ? (0)

Anonymous Coward | more than 2 years ago | (#40472525)

why would anybody run HP-UX on HP's x86 hardware? What would stop me from running it on Dell, IBM, et al x86 servers? I suppose it could be a contractual obligation, but that might be a slippery slope. Also, HP-UX is not the only operating system that runs on Itanium. Windows Server runs on it (or at least it used to), and VMS runs on it. I assume virtually nobody is running Windows on Itanium, but I know for a fact there are people running VMS on Itanium. And while there has been some success in emulating VMS on x86, it's not good enough for everybody.

Re:Why exactly ? (1)

Anonymous Coward | more than 2 years ago | (#40472591)

HP-UX doesn't run on x86. It runs on PA-RISC and Itanium.

Re:Why exactly ? (0)

Anonymous Coward | more than 2 years ago | (#40472815)

check the comment to which I replied. the parent said:

"Why dont they simply cut n run, and/or move HP-UX to x86 as they've already proven to themselves that it can run on x86."

My point was HP hurts themselves if they release HP-UX for x86.

Re:Why exactly ? (0)

Anonymous Coward | more than 2 years ago | (#40473659)

I wonder if they already have it internally. I remember a firmware update for an older HP print server that switched supposedly from Linux to HP-UX for the underlying os on it. That print server was based on a 486 CPU.

Re:Why exactly ? (3, Insightful)

Anonymous Coward | more than 2 years ago | (#40472543)

Considering that ditching Itanic would mean porting HP-UX, NonStop OS, and VMS, it would be a non-trivial exercise. First, you have to port the OS. Then you have to either write a compatibility layer (similar to Apple's Rosetta), or port all of the applications, including those you didn't write yourself. Given the tendency of business to not want to spend more money than they have to, the former is more likely in the short term.

There's also the political issues involved with abandoning a significant hardware lineup. Sure, there was a seven year overlap between Itanic and PA-RISC on the HP-UX side, but no customer is going to be happy about such a move. Compare with IBM; they've been doing POWER for two decades, and no end in sight. Or Oracle and SPARC, for that matter.

In my blunt opinion, if HP tries to move HP-UX to x86, it'll mark a major marketing campaign from IBM and Oracle about a company that won't stand behind its hardware, and they will bleed (more) marketshare.

This whole thing is not going to end well.

Re:Why exactly ? (0)

Anonymous Coward | more than 2 years ago | (#40472819)

They already ported HP-UX from RISC to Itanium and VMS from Alpha to Itanium... you'd think they'd have some practice.

Re:Why exactly ? (3, Interesting)

UnknowingFool (672806) | more than 2 years ago | (#40472557)

Itanium servers are very profitable for HP and they don't really have any competition in this server space. It is very expensive for Intel as they don't sell enough of them to justify the R&D and support. Intel would like to drop it but for HP. Oracle would like to drop it for support costs. Probably HP wants to keep their profits even though it costs money for everyone else.

Re:Why exactly ? (4, Insightful)

jedidiah (1196) | more than 2 years ago | (#40472677)

Are you kidding? Sparc and PPC are plenty of competition for HP in the server space. If anything, HP/UX has always been an ugly redheaded stepchild when it comes to Oracle support.

If you're running HP, you're already trying to smash a square peg into a round hole here.

That said: Oracle should still be held to any contracts it made.

Re:Why exactly ? (1)

bmo (77928) | more than 2 years ago | (#40472795)

FYI, SPARC and PPC don't run VMS, which is what this is all about.

--
BMO

Re:Why exactly ? (1)

UnknowingFool (672806) | more than 2 years ago | (#40477723)

The last time I checked Sun and PPC do not run Itanium. If you want Itanium, HP is really the only company. As such HP charges gads of money.

Re:Why exactly ? (1)

DragonWriter (970822) | more than 2 years ago | (#40473377)

Can somone opine as to why exactly HP is doing this ?

Money, mostly.

What do they hope to get ?

See previous response. (Both in damages, and by not losing their existing Itanium-based business.)

Why dont they simply cut n run, and/or move HP-UX to x86 as they've already proven to themselves that it can run on x86.

They will, but that doesn't recover costs they expended in reliance on the agreement they had with Oracle to maintain support on Itanium, so they still want to win the case.

I mean, even if they win everything (damages and an order to Oracle to continue support their DB on Itanium), they are going to know that Oracle is going to do the bare minimum to not be held in violation of the order, that that might soften the decline of Itanium a bit and provide some funds that can be used either to soften it a bit more or to underwrite a pivot off of it, but it won't make Itanium viable in the long run.

Re:Why exactly ? (1)

sjames (1099) | more than 2 years ago | (#40473639)

They've got Itanium, a brand new pair of dead whale kickin' boots and the world looks like a beach?

four /billion/ ?! (0)

Anonymous Coward | more than 2 years ago | (#40472401)

Okay, there's no way that product line is worth anything like that.

I guess you could argue that the damage to the company could push it over the edge. Hard to keep a straight face, but that's not required in court. But regardless that would be HP's problem and not something (I hope) you could extend to a supplier.

And yeah lawyers are expensive but still not that bad yet.

Is there no way to have batshit-level unrealistic damage demands considered contempt of court? I don't like that we're paying for this. (Both in coin, and tying up our limited court resources.)

Re:four /billion/ ?! (2)

Sir_Sri (199544) | more than 2 years ago | (#40473037)

Depends how they calculate 4 billion dollars. If their entire itanium business reseted on Oracle continuing support and the itanium business was 4 billion dollars then it's not crazy.

It seems crazy. But this is a lot of expensive stuff we're talking about. The hardware is cheap, it's the software, and importantly the data in the software that matters, and that's big money. Migrating it all to another setup could be expensive, replacing all the hardware is expensive etc.

Get Apple involved (1)

Anonymous Coward | more than 2 years ago | (#40472427)

An Mac Book Itanium would sell more units than anyone could produce.

Re:Get Apple involved (1)

pkinetics (549289) | more than 2 years ago | (#40474567)

wouldn't that be Apple iTanium?

Re:Get Apple involved (0)

Anonymous Coward | more than 2 years ago | (#40475711)

It would sell more heat sink units than anyone could product.

HP Itanium Support (1)

joelsherrill (132624) | more than 2 years ago | (#40472489)

I don't know what HP's plans were BEFORE Oracle dumped Itanium support but according to the HP-UX support maxtrix from February 2012, they will support some Itanium systems until 2018. I don't know if they killed any products early due to lack of Oracle software support but without Oracle support, I would bet there is every reason for many of the Itanium users to (1) cancel any planned Itanium purchases and (2) drop the existing ones. With them being taken out of service, HP loses revenue. It's a lot of money but it likely forced them to kill a product line early and encouraged existing more or less happy users to bail earlier than HP planned.

http://h71028.www7.hp.com/enterprise/downloads/public_hp-ux_systems_support.pdf?jumpid=ex_R1533_us/en/large/eb/go/hpuxservermatrix [hp.com]

Re:HP Itanium Support (2, Insightful)

gstrickler (920733) | more than 2 years ago | (#40472659)

And that's HP's problem, not Oracle's. HP didn't pay Oracle to develop, or maintain their software for HP systems. Oracle did it because they thought it was good business. All of the Itanium [wikipedia.org] sales projections over the years have been reduced by at least an order of magnitude. Now, with Itanic continuing to sink, Oracle has changed it's mind.

$4B is insane, it's nearly the entire Itanium market. And the claim that Oracle would agree to a contract that could cost them $4B as part of the settlement of a lawsuit over hiring a HP's ex-CEO, who resigned amid a scandal of his own making, a CEO who would almost certainly have been fired (and probably was told to retire), is simply absurd. No, HP is simply trying to keep a sinking ship afloat and trying to make Oracle stay and help bail water.

Re:HP Itanium Support (5, Informative)

bmo (77928) | more than 2 years ago | (#40472823)

>HP didn't pay Oracle to develop, or maintain their software for HP systems. Oracle did it because they thought it was good business.

Larry signed the contract. They're on the hook.

Just like David Boies signed a contract to prosecute SCO's lawsuits until the heat death of the universe, because he thought he was going to get a chunk of the 5 billion SCO was suing IBM for.

Greed leads to bad decisions. Who woulda thunk it.

--
BMO

Re:HP Itanium Support (2)

joelsherrill (132624) | more than 2 years ago | (#40472913)

We don't know what the contract said. We also don't know if HP subsidized the port of Oracle products to their Itanium line in exchange for some commitment.

$4bn is probably more money than HP thought it was worth but you have to have room to negotiate.

We can be pretty sure that dropping Oracle support did not help keep people using HP Itanium computers.

If Oracle violated a signed contract, then they deserve this. Otherwise, it is no more of a waste of court time than anything Apple has done.

Re:HP Itanium Support (1)

gstrickler (920733) | more than 2 years ago | (#40473073)

That's not exactly how contract law works.

First, we don't know if the contract specified any particular length of time, or just a general "We'll continue to develop..."

Second, in contract disputes, the intent and what is fair and equitable are as important, and often more important than the letter of the contract.

Third, there is a "would a reasonable person willingly agree to such terms" guideline applied in resolving contract disputes.

Fourth, we don't know if the contract specified any penalties for breach of contract, and since neither side has said there were specified terms, it's reasonable to assume they were boilerplate "...may be liable for damages...", damages which HP must prove are due solely or predominantly to Oracle's decision, and not caused by HP or other market forces.

IANAL, but your blanket statement

Larry signed the contract. They're on the hook.

is simply not correct.

Re:HP Itanium Support (3, Insightful)

bmo (77928) | more than 2 years ago | (#40473285)

Oracle only gets out of the contract for development if the clause on it is unconscionable or otherwise unenforceable.

Unlike your average joe, Oracle has lawyers that they pay to go over this stuff and pick out and cross out the unenforceable and unconscionable stuff for revision before signing.

IANAL, but I trust Oracle hires good lawyers.

--
BMO

Re:HP Itanium Support (1)

gstrickler (920733) | more than 2 years ago | (#40473415)

I would say a $4B penalty claim in such a case is unconscionable, and not foreseeable. Even HP's low end claim of $500M is unconscionable given that the market never developed and the Intel itself has indicated the CPU doesn't have much of a future.

Re:HP Itanium Support (3, Insightful)

bmo (77928) | more than 2 years ago | (#40473635)

>I would say a $4B penalty claim in such a case is unconscionable, and not foreseeable.

No. That's not how it works. It's not whether the penalty is unconscionable. That's just HP asking for relief. They can ask for any number they want. But in cases like this, you always ask for more than what you need because it only gets adjusted down anyway.

You only get out of it if *the clause in the contract itself* is unconscionable or otherwise unenforceable. According to HP, Oracle signed a contract saying that Oracle would continue to support Itanium. HP is telling the court that to make them whole for Oracle to break the contract, it would be 4 billion to call it even, because that's what the projected damage would be, according to HP.

Whether the court agrees with that amount or not, the court has to first determine whether the clause itself was unconscionable or otherwise unenforceable. If it's a valid clause, it's just a matter of determining how much it would be to make HP "whole" for Oracle breaking the contract.

Proving the clause is unconscionable or otherwise unenforceable is a pretty high hurdle for Oracle to clear, since their lawyers are experienced in handling contracts like this and they should have known before signing that it was unconscionable. Proving this means that their lawyers were incompetent at the time of signing. Not proving it means that their current lawyers are incompetent.

THE CONTRACT IS INVALID BECAUSE I WAS DRUNK, YOUR HONOUR.

--
BMO

Re:HP Itanium Support (1)

gstrickler (920733) | more than 2 years ago | (#40473701)

Obviously my sarcasm tag didn't work.

Re:HP Itanium Support (1)

bmo (77928) | more than 2 years ago | (#40473857)

Obviously.

I'll go get my meter checked.

--
BMO

Re:HP Itanium Support (1)

gstrickler (920733) | more than 2 years ago | (#40475729)

And I'll have my encoder inspected. It seems to have missed a few times lately.

Itanium 3 is a decent CPU (4, Interesting)

jd (1658) | more than 2 years ago | (#40472547)

The original Itanium was a disaster, the second generation was what the first should have been but wasn't, the third actually looks very respectable. Intel would be stupid to eliminate a product they've actually got functional.

And for high-end use, the Itanium is a genuinely useful CPU. Because the performance of a cluster is a function of the communication delays, very high-end clusters WANT to have very high-end CPUs. You can only do so much with piles of PCs before the inefficiencies due to (a) distance and (b) an inefficient architecture really set in. There's also (c) - a crap instruction set - but the Itanium doesn't help much there because although it is somewhat better, nobody has built a particularly good compiler for it yet. Optimization on the Itanium remains a challenge.

Admittedly, it's not the design I would have chosen - I far prefer many of the design decisions made in the Inmos Transputer and the Intel iWarp, since those were designed specifically for the purpose of clustering and started from that position. I also prefer the elegance of the MIPS64 instruction set over the unnecessary burden of anything Intel has done, but again I'm in the minority. I'd also have threaded compute elements and produced virtual cores, rather than threaded instructions on physical cores, since threading the compute elements would allow you to distribute the heat better, wouldn't prevent you accessing elements that are wholly independent of those in use and would reduce unnecessary swapping. But what do I know, I've only been observing what actually works vs what the customers want for 35 years. Customers are just as stupid as beancounters and pointy-haired bosses.

Re:Itanium 3 is a decent CPU (4, Interesting)

multiplexo (27356) | more than 2 years ago | (#40472757)

Yeah, it's too bad to see it go. I ran an SGI Altix 3000 at a previous job and it was a screaming box, very elegantly designed and designed for massive clustering. NASA had a cluster of 10,240 Itaniums using the SGI Altix architecture, and it wasn't a weird, one-off hack. You could have duplicated it if you had the money and the space. I think that not having an affordable way for hobbyists to build their own Itanium workstations really hurt Intel. If Intel, or someone else, had come out with an affordable motherboard for building a single or dual CPU system more people could have built their own systems, as they do with x86, and seen what the chip was capable of. What Intel has accomplished with the x86 is impressive, but how long can it go on?

Re:Itanium 3 is a decent CPU (3, Interesting)

Anonymous Coward | more than 2 years ago | (#40472899)

When I was at Sun, and Sun's first Opteron boxes came out, we were told that we lost that NASA contract to SGI because intel paid SGI to give the cluster away for free to NASA (for the publicity) and only to charge for the support...

The itanic was never that great, even compared to the first Opterons (which used less power and ran cooler too). It was effectively a large DSP and no one (no engineers) in the industry outside of intel took it seriously. It was only ever going to be any good at highly-predictable numerical workloads due to the limitations of the instruction set. The "good" compilers that it was to require would never appear since you can't write a compiler that can predict the future.

The other benefits of itanic such as the large caches, multiple cores, wide registers and fast interconnects were already available on other CPUs... including the Alpha it killed and the Opteron/AMD64 that we have today.

Re:Itanium 3 is a decent CPU (2)

Teresita (982888) | more than 2 years ago | (#40474273)

NASA had a cluster of 10,240 Itaniums using the SGI Altix architecture, and it wasn't a weird, one-off hack. They used it as a CGI render farm for their fake Mars rover program.

Re:Itanium 3 is a decent CPU (1)

GGardner (97375) | more than 2 years ago | (#40473301)

And for high-end use, the Itanium is a genuinely useful CPU. Because the performance of a cluster is a function of the communication delays, very high-end clusters WANT to have very high-end CPUs.

Note the above is certainly true for high-end HPC clusters, but running large Oracle databases on those kinds of machines seems kind of expensive for the performance you get. For Oracle (and other databases), the high-thread count Sparc T-3 / T-4 kinds of processors will give you much better performance at lower cost. Of the few ia-64 installations, I bet most are floating-point heavy HPC clusters, I wonder how many are running Oracle or VMS and "business" workloads.

But what do I know, I've only been observing what actually works vs what the customers want for 35 years

Of course, if customers actually wanted Oracle on Itanium, there wouldn't have been a lawsuit...

Re:Itanium 3 is a decent CPU (1)

jd (1658) | more than 2 years ago | (#40473931)

Well, yeah. Anyone doing FPU-intensive work on a database (especially an Oracle one) is in the wrong line of business and probably needs considerable therapy. Certainly for database work, you WANT to have as much of the CPU dedicated to running large numbers of relatively basic threads - something the UltraSPARCs and successors do BRILLIANTLY. Hell, if you scaled a StrongARM to handle that many cores and threads, I'd take that over a number-crunching chip for database work.

Oracle RAC is the clustered version of their database (IIRC) but to take advantage of it you need a network card that'll do kernel bypass. By definition, that doesn't involve the CPU beyond the CPU allowing something else to control the bus. After that, it's ALL down to bus speeds and how many independent channels you can run.

Re:Itanium 3 is a decent CPU (1)

Richard_J_N (631241) | more than 2 years ago | (#40473359)

I don't understand Oracle's position here.Surely all they have to do is compile the latest release of their software with the relevant compiler options?

Re:Itanium 3 is a decent CPU (1)

slew (2918) | more than 2 years ago | (#40473473)

Intel would be stupid to eliminate a product they've actually got functional.

Companies do this all the time. When the expected revenue doesn't meet the support costs plus the opportunity cost, it's better to eliminate the product than continue to support it. Just getting something functional is a business 101's definition of a "sunk-cost".

I far prefer many of the design decisions made in the Inmos Transputer and the Intel iWarp

Interesting you picked those two. I was working at Inmos when CMU came up with the iWarp If the transputer were to be redone from scratch, it probably would have looked very similar to iWarp, but then I thought to myself, aren't people supposed to avoid these mistakes ;^)

But I digress. Today Intel is betting on MIC instead of Itanium.

I'd also have threaded compute elements and produced virtual cores, rather than threaded instructions on physical cores, since threading the compute elements would allow you to distribute the heat better, wouldn't prevent you accessing elements that are wholly independent of those in use and would reduce unnecessary swapping.

AMD's Bulldozer is an example of partially virtualizing the cores (basically two threads share 2 cores) and it doesn't seem to be that great. I'm not so sure that it necessarily suggests that this strategy has much legs for problems that are solved in "big-iron" computers anymore than low latency access to high-latency serial async messaging in the TP or iW case. For some small class of problems it's a slam dunk, for the general case, not so much...

Re:Itanium 3 is a decent CPU (2)

cheesybagel (670288) | more than 2 years ago | (#40473769)

Bulldozer basically does what processors like Niagara did before i.e. it shares the FPU among cores to reduce die space. It works marvelously for web servers or java applications servers i.e. multi-threaded integer applications but sucks for everything else.

Re:Itanium 3 is a decent CPU (2)

jd (1658) | more than 2 years ago | (#40474465)

To be fair to Inmos, the Transputer was really only competing with big iron - no, I do not take the Transputer module for the Amiga seriously.

Even if it had simply evolved into a networking chip (rather than a DVD processor, which is what actually happened), the modern world spends a couple of grand on individual high-end SCI interconnects (which are essentially ptp serial links), which are distributed over an entire PCI card rather than being a system-on-a-chip. Being able to replace four such boards with one wafer would seem extremely desirable. It should certainly be faster, since all variables are the same bar the one of the Transputer being on a single die. The same goes for the iWarp.

If you want to compare apples with something that is still vaguely apple-shaped and within the category of fruit-like, then look at the sheer difficulty anyone has had superscaling the Intel processors. SGI managed it and went bust. Other than that, not many vendors have ever gone beyond sixteen-way. In comparison, Cambridge Computers were chucking thousands of Transputers into a single crate. IIRC, this was a Clive "people want washing-machine-motor cars" Sinclair operation we're talking about, which means it has to have been easy enough for him to comprehend. That's.... ....not terribly demanding on the gray-cells. (He's in MENSA, but I can only imagine it's because they were using a Sinclair-made watch to time the test.)

Agreed, the actual T400 and T800 had architectural flaws and the iWarp had problems, but modern clusters are merely the same concept done over multiple motherboards with MPI-2.1's ghastly API to handle most of the data transfer. There's no intrinsic reason why a SoC version could not be the equal or superior, it merely requires a slightly larger die and a slightly better design.

Ok, what about the general case for virtual cores? Well, the reasoning is that applications place different demands on the CPU. Skype, Tor and Firefox are not equal in their needs, for example. People tend to run a mishmash of programs, it's actually quite rare to run many programs that happen to have identical needs. So rare even in games consoles that IBM opted for a hybrid architecture in their Cell design rather than try to make multiple cores that could do everything well.

In big iron, people tend to write SIMD code because you can't do MIMD on a heterogeneous cluster very well. With virtual cores, it would be the other way round - you'd be able to do MIMD far better than SIMD, since you could divide the processor elements up into more groups. So the question then becomes whether real-world problems are SIMD or MIMD. Since both still fall under the category of Turing-Complete, either can do the problems of both, so it's no use asking what people use. The question is what would be more efficient. Given that people DO use MPI and PVM, I'm less than convinced that people have looked into efficiency when scaling up will work.

Re:Itanium 3 is a decent CPU (1)

Anonymous Coward | more than 2 years ago | (#40474823)

I don't recall many (any?) periods in the lifespan of the Itanium where it was the best performing CPU, or gave the most bang per buck. It may have been fastest for a few very parallel scalable tasks, but very parallel scalable tasks have a habit of being able to be done faster on a cluster of cheap x86 nodes.

If you want a very high end CPU, you'd go for IBM POWER. If you didn't, you went for AMD64. Furthermore the Itanic was merely a CPU, whereas IBM POWER had a lot around it. HP did provide systems, but in the market they were left with, IBM could provide what HP did and more.

So the options for performance back then were: IBM POWER (fastest but more locked in, most expensive), Intel/AMD (not as fast but less locked in and less expensive). SPARC was losing despite Fujitsu's efforts (Sun was busy creating CPUs that were weak and obsolete by their release day). The Itanic could not beat the x86 except for a few specific tasks/areas, had lock in, was expensive, and there wasn't much existing code around that it could run fast.

Apparently when the Itanic was slow it was often really slow. Whereas the x86 would perform better on unoptimized or crap code (the x86 ISA has been encouraging crap code for years, so Intel/AMD need their CPUs to cope well with the ocean of crap out there).

quandary! (2)

roc97007 (608802) | more than 2 years ago | (#40472553)

I can't stand either company, for different reasons, and have absolutely no interest in Itanium. I have a hard time picking someone to root for in this... I guess I'll have to go with HP. Go, HP! Only because (a) it's entertaining, and (b) it causes problems for Oracle.

If Oracle counter-sues, I can always root for Oracle.

Re:quandary! (2)

jamesh (87723) | more than 2 years ago | (#40474217)

I can't stand either company, for different reasons, and have absolutely no interest in Itanium. I have a hard time picking someone to root for in this... I guess I'll have to go with HP. Go, HP! Only because (a) it's entertaining, and (b) it causes problems for Oracle.

If Oracle counter-sues, I can always root for Oracle.

Hopefully the shareholders realise that every dollar that goes to the various law firms is a dollar that won't be part of the profit pool they get dividends from. Sure, $4 billion is a huge loss or gain for the parties in question but i don't think anyone believes that $4bn is ever actually going to change hands.

If I was placing a bet on anyone, i'd be putting my money on the lawyers. That would be a sure bet.

Re:quandary! (1)

roc97007 (608802) | more than 2 years ago | (#40475463)

> Hopefully the shareholders realise that every dollar that goes to the various law firms is a dollar that won't be part of the profit pool they get dividends from.

Bonus!

HP v. Oracle (2)

Guppy06 (410832) | more than 2 years ago | (#40472599)

Any chance that they will both lose?

A boy can dream...

Try MySQL (3, Funny)

drooling-dog (189103) | more than 2 years ago | (#40472643)

If Oracle doesn't want to support it on your platform, you can do it yourself. For less than $4 billion, anyway...

Re:Try MySQL (0)

Anonymous Coward | more than 2 years ago | (#40473393)

If Oracle doesn't want to support it on your platform, you can do it yourself. For less than $4 billion, anyway...

At the risk of feeding a troll...

Businesses who use Oracle on VMS have performance and capacity needs that toy databases can't meet. Replacing oracle with mysql is like replacing a an airplane with a skateboard.

Re:Try MySQL (3, Funny)

laffer1 (701823) | more than 2 years ago | (#40473675)

To be fair, it would be a cluster of skateboards.

HP-UX ancient OS (1)

Anonymous Coward | more than 2 years ago | (#40472705)

Since HP hasn't really been spending any cash to innovate with HP-UX, and still has many what I'd consider bugs in the OS, there's really no reason for anyone with a modern OS mindset to use HP-UX. It's only for legacy apps in my opinion. Legacy apps don't need updates or upgrades - they need to continue to run until the wheels fall off.

I see HP's Itanium line going the way of their PA-RISC line of HP-9000 servers, as well as MPE. PA-RISC at least had something going for it - as did the DEC-Alpha systems they pretty well killed.

Sorry HP, but you are a has-been nobody today.

Re:HP-UX ancient OS (0)

Anonymous Coward | more than 2 years ago | (#40472745)

in case anyone is interested - bug #1 - scsi driver written so that cannot have more than 7 scsi devices (including controller) per scsi target....

For disks, it automatically increments the target ID once you hit the 8th device, for tapes, you have to manually zone the tape drives as a separate mapping to get new targets assigned. (that's as of HP-UX 11i v3, the last rev I tried to use)

If they can't get basic things like scsi disk and tape interfaces updated to something remotely modern, I shudder to think how it would handle things like iscsi targets.

It's a sad sad state of affairs

you are full of shit (1)

rubycodez (864176) | more than 2 years ago | (#40475649)

wiht U320 devices, there are 15 allowed devices per chain besides the adapter, and under HP/UX the scsi priority is 7, 6, 5, 4, 3, 2, 1,0, 15, 14, 13, 12, 11, 10, 9, and 8. Multiple HBA can be on the same bus and those are assigned the highest priority numbers.

EnterpriseDB (3, Interesting)

Defiler (1693) | more than 2 years ago | (#40472869)

It seems to me that HP would be better off sinking this money into contributions to PostgreSQL / EnterpriseDB; it already offers a ton of Oracle compatibility, and runs on HP-UX: http://enterprisedb.com/products-services-training/products/postgres-plus-advanced-server/advanced-server-oracle-features [enterprisedb.com]

Re:EnterpriseDB (1)

Anonymous Coward | more than 2 years ago | (#40473425)

It seems to me that HP would be better off sinking this money into contributions to PostgreSQL / EnterpriseDB;

Sinking what money? HP is suing oracle. In other words, HP wants oracle to pay a penalty because they are breaching their contract.

Re:EnterpriseDB (1)

Defiler (1693) | more than 2 years ago | (#40473557)

I mean the money/effort spent on this litigation.

Re:EnterpriseDB (0)

Anonymous Coward | more than 2 years ago | (#40473751)

for whatever it is worth, I have heard more than one HP big-wig say they don't want a database product. Admittedly I haven't heard one say it since Meg took over.

Re:EnterpriseDB (1)

Billly Gates (198444) | more than 2 years ago | (#40473903)

HP would have made A HELL OF ALOT MORE money than offering support and cheap lintel servers running a free database. Yes HP has been mismanaged for years but Oracle screwed them and violated their contract and should pay HP up.

$20,000 per server is what HP probably would have made as these systems are big iron and HP would sell their OS VMS, HP UX, and Nonstop in addition to enterprise software like openview. A $999 HP lintel x86 server doesn't offer that much. Even their more enterprise higher end lintel and wintel servers do not have the margins with the extra software.

It also would cost billions to port their operating systems to x86 and it might not even be worth it now as their customers decided or are starting to replace their platforms with lintel and wintel solutions from competitors and in house. Could HP-UX survive on x86 now? I don't know.

I hope Oracle pays them back big as they did this to screw HP over to increase their Solaris systems with their database stack.

Make oracle pay (0)

Anonymous Coward | more than 2 years ago | (#40473025)

Make oracle pay. Larry Ellison doesn't need to buy Lana'i.

HP-UX on x86 (1)

randyapenguin (942577) | more than 2 years ago | (#40473175)

I've read that the Itanium has hardware partitioning and redundancy features that the x86 still doesn't have. If lives are on the on the line and cost was not a barrier it would be easy choice to make. If Oracle won't support HPUX on Itanium, why does anyone think they'll support it on HPUx86. Oracle is being a bitch and has also increased licensing costs so many customers have been caught out by this. This is a big pissing match with customers losing out.

Re:HP-UX on x86 (1)

cheesybagel (670288) | more than 2 years ago | (#40473885)

They still support SPARC and POWER so who cares... A lot of HP's old clients already saw the writing on the wall a long time ago and switched to one of those. Meanwhile x86 keeps getting more features so eventually that won't matter.

Re:HP-UX on x86 (1)

rubycodez (864176) | more than 2 years ago | (#40475663)

x86 does it another way, the modern way is to virtualize on x86 blades, HP already had x86 port of HP/UX that can run on vmware ESXi on blades, google project "Odyssey"

Quality software if they force support ? (1)

MarkTina (611072) | more than 2 years ago | (#40473583)

If Oracle get forced to support Itanium I wonder what quality software they will actually make ?

thamks (0)

Anonymous Coward | more than 2 years ago | (#40474423)

http://multx.ru

Micsoft will buy Oracle (1)

SvenLee (2624751) | more than 2 years ago | (#40475479)

What a slap in the face. Hp or Microsoft?
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?