Slashdot Mirror
Oracle Ordered To Pay $3B Damages To HP (bbc.com)
Oracle has been ordered to pay HP $3 billion in damages by a California jury over HP's claim that Oracle reneged on a deal to support HP computer servers running on Itanium chips from Intel. Oracle said it will appeal. BBC reports:The court battle over the contract was settled in 2012 but the damages HPE was due have only now been agreed. HP was split into two in 2015 with HPE taking over the running of its servers and services business. In court, HPE argued that although the 2012 legal judgement meant Oracle had resumed making software for the powerful chips, its business had suffered harm. It argued that Oracle took the decision in 2011 to stop supporting Itanium in a bid to get customers to move to hardware made by Sun -- a hardware firm owned by Oracle. Oracle said that its decision in 2011 was driven by a realisation that Itanium was coming to the end of its life. It also argued that the contract it signed never obliged it to keep producing software in perpetuity. Intel stopped making Itanium chips in late 2012 and many companies that used servers built around them have now moved to more powerful processors.
/s
If Oracle can be made to pay fines for it's ability to push the limits of capitalism into the grey area of criminal activity, then this is a good thing.
I once work for Oracle, after 6 months, I went looking for a new job. Sun hired me. When Oracle acquired Sun, I quit working at Sun/Oracle.
Oracle on the Sales, Marketing, Management side is just people that have no moral compass other than me, me, me.
The Engineers, I am disappointed in, but they also are me, me, me.
So why did it take 2 years?
The judge has been too patient and lenient.
Wasted time and delayed punishment usually result in the beast/child not associating negative consequences with the
causal action...
Spank them! Fast, before they can deal for justice.
Never stop, ever.
You are the wind beneath my wings app.
The Sun engineers who mostly quit en masse during the acquisition?
Unity? Screw that: XFCE. Slashdot Beta? Screw that: SoylentNews. Australis? Screw that: Pale Moon. UX developers DIAF
Sure, but don't forget that Sun bit the big one because their sales people would not sell x86/64 products because they got a bigger margin on the Enterprise server gear.
Customers would not buy a ton of big iron when a little big iron and a shit-ton of cheap x86/64 gear would fit the job just as well and cost a lot less money
Sun sales execs bear just as much responsibility as anybody for the situation that you faced at Sun
I honestly don't care that oracle has to bleed, for a number of reasons. In fact perhaps it is deserved in some way. But I'm actually miffed hp is getting the proceeds.
PA-RISC and Alpha AXP are missed. itanic is not.
According to wikipedia, "According to Gartner Inc., the total number of Itanium servers (not processors) sold by all vendors in 2007, was about 55,000".
Unless a zillion Itanium processors (not servers) were sold, how do the damages add up to 3B?
Between this, losing the Java case, and the whistle being blown on their cooked accounting books, it really seems like Oracle's legal department's about to get tied to a pole and whipped until bloody by Ellison until they've expiated their failures.
The recent "Subway" supercomputer cluster is supposedly based on an Alpha 21164 design. https://en.m.wikipedia.org/wik...
Sun's sales organization may have been dysfunctional, but that isn't what killed them. Sun had a broken business model. They depended on selling high margin hardware (e.g. original SUN workstations and then SPARC servers), and that market was disappearing. X86 caught up. Sun couldn't compete with commodity hardware that was "almost as good" as SPARC, but cost half as much. Also, Sun couldn't support their cost structure by selling X86 servers -- it wouldn't have mattered if the sales org was totally dedicated to X86 and sold tons of them.
HP owned Alpha and PA RISC, both respectable established architectures. Alpha in particular gets a lot of praise.
So why did HP dump them for Itanium? Simple answer: Because it was going to cost too much to stay competitive.
HP couldn't afford to keep designing their own processors. Designing high end processors is expensive, and the costs were escalating. HP would need to spend hundreds of millions (if not over a billion) dollars on R&D. And the spending never stops. Processor designs have a limited lifespan -- you've only got a few years before the competition leapfrogs you. HP realized they couldn't sell enough servers to cover those costs.
Itanium turned out to be a complete disaster. "cluster fuck" doesn't even begin to cover it. But that doesn't mean HP was wrong to ditch Alpha and buy processors from a processor company. You need to make decisions about these things years in advance. We can look back now and say that Itanium was an eipc failure, and wonder how HP could have been that dumb. But that's not how it works. At the time the decision had to be made, it wasn't unreasonable.
P.S. Despite being direct competition to Itanium, HP was an early supporter of the AMD x64 architecture. HP sold lots of servers with AMD Opteron and Intel Xeon processors, and helped push commodity servers to 64 bits.
P.P.S. I don't know anything about the legal merit of HP's case against Oracle. I do know that Larry Ellison and Oracle deserve to get kicked in the nuts as often as possible. So I'm happy they have to pay $3B to HP.
In one sense I like that a lot. In another, not so much.
The problems are that all the documentation will be in Chinese, and for all their totalitarian control they have quality and other control issues that mean there may well be backdoors in there. Even the USG plays dirty tricks in that space, and they're supposed to have some compunction about diddling hardware, where the Chinese clearly have none at all, and manage to be a lot less open about it.
I would probably sooner trust a SPARC64 VIIIfx ("K computer" cpu) than this one. So in one sense this SW-3 is pretty cool indeed, but then there's the trust issues and the language barrier. And, of course, much like the VIIIfx, the cpu likely won't be available for server or desktop use, n'mind laptop use.
On another note, it might be interesting to see what'd happen if we could run this cpu on the K computer's Tofu interconnect, as that was what allowed the K computer to run at very high efficiency.
It does answer a question that I've been wondering about, whether alpha would still perform when (re)implemented on a modern process. The answer looks to be "it'd still kick arse", so that's nice to know. Too bad nobody will actually get to use it except a few academics.
Intel stopped selling Itanium chips in 2012, but somehow Oracle is still supposed to keep supporting these chips forever? HPE is evil and trying to recoup treasures from a sunken ship.