back to article HP started then spiked HP-UX on x86 project

As part of the ongoing lawsuit about whether or not Oracle had committed itself to supporting its software on Hewlett-Packard's Itanium-based servers, the software giant did a core dump of very interesting documents that show what many of us suspected: that HP did indeed mull acquiring the Sparc/Solaris business and that HP did …

COMMENTS

This topic is closed for new posts.
  1. Anonymous Coward
    Mushroom

    Wow, HP realy are slinging there own mud about

    HP are starting to go a bit emo over this, so what they started then canned porting HPUX onto Itanium, IBM ported AIX to Itanium, run for a release and then dropped it like a ball and that was around 10 years ago so how slow can HP be. Saying they delayed as they were thinking about buying Sun is frankkly a load of bollock and if true is such bad managment for being so late to porting HPUX in the first place. No wonder Oracle was like - sod supporting your dwinderling platform. IBM was never going to buy Sun and anybody in the industry could of seen that was the case, IBM dont like being in a position of being a monoply due to historical bad press and since Microsoft and then others came along have been able to get back to work without being brow beaten every other year as they used to be. They did after all invent having lawyers in IT pretty much, so know what they can and cant get away with and prefer to stay under the radar.

    Realy starting to look like HP are going to throw out all there toys from there pram and wonder down the line were it went wrong and still end up blaming others.

    1. Anonymous Coward
      Anonymous Coward

      Re: Wow, HP realy are slinging there own mud about

      It is not the porting to x86 that is the issue. HP can port HP-UX, VMS, NonStop to x86 from a technical perspective. The issue is that the software providers have said that they do not have any interest in adopting HP-UX on x86 because of the development effort on their teams and the fact that HP-UX on x86 is likely to be a very small portion of the market share. They don't want to re-create a port (for the third time) for HP-UX if only a handful of people will be interested. They basically told HP that HP-UX is going to have to stay on Itanium if they want support.

      1. Anonymous Coward
        Coat

        Re: Wow, HP realy are slinging there own mud about

        Don't you mean second time? (As far as I know HP-UX started out on PA-RISC and then went directly to Itanium.) While if OpenVMS would be ported to x86 it would be the third time.

        1. Anonymous Coward
          Anonymous Coward

          Re: Wow, HP realy are slinging there own mud about

          HP-UX ran on M68K CPUs in the 80s. The port to PA-RISC was easily accepted by customers since PA-RISC performance was a massive step up (it, along with DEC Alpha, held the performance crown for much of the 90s)

        2. Anonymous Coward
          Anonymous Coward

          Re: Wow, HP realy are slinging there own mud about

          Yes, the x86 port would be the third (although, according to DougS, the fourth). Three ports in five years would fragment the user base... and the ISVs don't want to support an x86 port because the odds are the most people would just use Linux if they went to x86. The market for an OVMS and HP-UX on x86 port would be small. Not worth the time of the ISVs.

    2. Kebabbert

      Re: Wow, HP realy are slinging there own mud about

      @PXG

      "...IBM was never going to buy Sun and anybody in the industry could of seen that was the case, IBM dont like being in a position of being a monoply due to historical bad press..."

      What? IBM loves monopoly, dont you know that? IBM fights very hard against any company that tries to break IBM monopoly on the Mainframes. IBM is sentenced to billion dollar fines, or buys the company. In case you didnt knew, IBM earns lot of money from their Mainframes. It is really big business. In practice, IBM has monopoly on Mainframes and fights to uphold it.

      Every company strives to a big market share as possible. The best is monopoly, and the ultimate goal. That is the reason we have laws against monopoly - it is too lucrative. IBM not wanting monopoly? You should read basic courses in business or economy.

      .

      .

      Of course IBM wanted to buy Sun. They were far in negotations when Oracle snapped Sun in front of IBM. IBM would have killed Solaris and SPARC, and offered a cheap migration path to POWER/AIX. Solaris is the most common Unix out there, so IBM would have a big market. Then only HP-UX would have been left. And IBM would soon have reached monopoly on Unix too. IBM could then increase prices on Unix as they liked, and have monopoly on both Mainframes and on Unix servers. Left would Linux be, for lowend. IBM would have monopoly on all highend servers. And highend is high margin. Low end is low margin. Golden opportunity. Very very very lucrative. IBM must have been really pissed when Oracle snatched Sun.

      You are totally off, believing no company wants to increase its market share as much as possible. AIX is not most common Unix, but IBM wants it to be, no matter what you think. Monopoly is the ultimate goal. Or close to monopoly.

      1. Anonymous Coward
        Anonymous Coward

        Re: Wow, HP realy are slinging there own mud about

        @Kebabbert "What? IBM loves monopoly, dont you know that?" Ok, what I was trying to say and failed was IBM don't like to be seen as a monoply, Yes they will take what they can, but they have got alot lot better in making them not appear to be one and in many area's stepped away from the limelight whilst still contolling alot behind the scene.

        I wasn't aware of IBM trying to buy sun and had they wanted to they would of had to break up parts and sell of other parts - tere is only so much donating to open source that wil plicate. As for porting customers over to POWER/AIX, probably I know I was consulted on a few migrations over 15 years ago regarding such ports and I can only assume they have increased beyong how to convert CSH to KSH. IBM give out AIX for free, alas you need to buy there kit to run it on so they still make there income one way or another.

        So once again sorry I was trying to say and failed badly in saying that IBM whislt not likeing to be seen as a monoply as it incures wrath, will take what they can, though wont make a song and a dance about it and try to avoid limelight unless its some R&D project they hope to sell as a new product and in that nobody argues about monoplies on new markets.

        1. Matt Bryant Silver badge
          Facepalm

          Re: Re: Wow, HP realy are slinging there own mud about

          "....what I was trying to say and failed was IBM don't like to be seen as a monoply...." But IBM were the first company to make an overt move in talking a merger/buyout with Sun back in 2008, long before hp and Orcale got involved publicly. And IBM were looking to take the whole carcass, it seems they just weren't interested in paying that much for it. Which would have given them control of Java (snif-sniff - monopoly!), lots more patents to bash threats to their mainframe business (ubermonopoly), and MySQL (which they probably would have raped into some low-cost, proprietary DB2 alternative, with the aim of killing Oracle's DB bizz and gaingin another monopoly).

          1. Anonymous Coward
            Anonymous Coward

            Re: Wow, HP realy are slinging there own mud about

            Yes, IBM was legitimately interested in Sun, primarily for Java. Yes, the would have likely killed Sparc, as Oracle will eventually, because it is a dog of a chip that is years behind the curve and doesn't have the market share to support the billions in needed investment. IBM would have never killed Solaris. They would have added it as an additional OS to the Power line.

            Mainframe has 0.0% to do with the Sun acquisition. No one has replaced a mainframe with a bunch of T-Sparc severs, apples and oranges. Almost none of the tech is shared with the mainframe platform. The concern was that IBM would own the Unix market with both AIX and Solaris which would have attracted regulatory attention. Sun wanted IBM's contract to buy to include a clause about supporting the acquisition process even if any reg flags were raise about Unix market share control. IBM didn't want to battle this out in the courts, especially as they were taking over the Unix market anyway. Easier to beat them in the market than buy them. IBM and Oracle both wanted Java, Oracle really wanted MySQL as well, neither cared/cares all that much about Sun hardware... at least not since 1998. Oracle was trying to dump Sun hardware on HP, IBM, Dell... anyone who would take it. They just wanted Java and MySQL.

            IBM would not have turned MySQL into a proprietary OS. That would have destroyed the value of MySQL. They would have likely used it to attack Oracle and MS SQL. IBM is the primary commercial supporter of open source.

            IBM only ended up bidding 5 cents less per share than Oracle. The bids were about the same, but Oracle did not have the regulatory hang ups because they did not have a Unix platform and did not anticipate MySQL being such a major issue with the EU. The major difference was that Oracle agreed to go through with the acquisition regardless of regulatory interest and IBM did not want to sign a contract that would have forced them to battle out the Unix issue for years in the courts.

      2. Anonymous Coward
        Anonymous Coward

        Re: Wow, HP realy are slinging there own mud about

        Kebabbert, as a Sunshiner, you should be upset that IBM did not acquire Solaris. They would have ditched Sparc, put Solaris on Power and Solaris would be in much better shape today than it is under Oracle as the "oh yeah, I guess we still sell that" product behind Exa.

  2. arrbee
    Meh

    I thought HP's case was that they had a binding agreement with Oracle to support itanium for x years and that Oracle broke that agreement ? In which case all this stuff is interesting but irrelevant.

    Maybe the suggested HP-UX port was killed to avoid demands to do the same for VMS.

    I don't know about HP-UX but the VMS user base was crying out for an i86 port many (many) years ago, as soon as Intel and/or AMD showed that the i86 wasn't a dead end - VMS having already suffered commercially from being on a minority platform it was fairly obvious what would happen.

    1. Anonymous Coward
      Anonymous Coward

      "Contract"

      The "contract" that HP claims to have which would entitle them to Oracle support on Itanium is the Mark Hurd press release. HP and Oracle issued a joint press release after the Hurd hire by Oracle and the resolution of the lawsuit which was basically a corporate hug (HP and Oracle will continue to support their joint customers, etc). HP is arguing that this press release was a contract, even though Oracle explicitly rejected the inclusion of language about Itanium in the PR, which contractually compels Oracle to support Itanium... presumably until HP says they can stop. In other words, there is no "contract" in the traditional sense, just a long shot on HP's part.

      1. Matt Bryant Silver badge
        Facepalm

        Re: "Contract"

        "The "contract" that HP claims to have which would entitle them to Oracle support on Itanium is the Mark Hurd press release...." And there you have it, right fromt eh mouth of a lwayer working on the hp legal team. Oh, no it's not, it's just that tired old Elmer again. In short, Wunderburp hasn't a clue as to which "contract" hp are referring to, but it's more likley to be the joint development agreement that preceded Hurd's arrival at hp.

        1. Anonymous Coward
          Anonymous Coward

          Re: "Contract"

          Enlighten me, Matt. What "contract" is HP referring to if not the Hurd press release? They have never mentioned any "joint development" contract in the legal filings other than the Hurd press release. Oracle has said that they "do not believe, nor do they think HP believes, that the Hurd press release was a contract for support." If HP has another contract, they have never mentioned it. Oracle is certainly under the impression that the Hurd PR is what they are debating. Every time HP has mentioned a "contractual obligation" it is very vague and always points back to the Hurd press release.

    2. Anonymous Coward
      Anonymous Coward

      VMS is done

      Even if HP could get VMS to x86 and anyone was interested in going through the migration effort to still be on VMS at the end of the day, Oracle has killed DEC Rdb, which they some how ended up owning, on all platforms. Discontinued the product.

      1. Anonymous Coward
        Anonymous Coward

        Re: VMS is done

        If I remember correctly; DEC Rdb was sold off to Oracle by R.Palmer, the man who almost destroyed IBM, got thrown out before he managed that and then succeded joining and completely destroying DEC.

        That said in the 25 years I worked with VMS (VAX, Alpha and 1 crappy Merced box), not once did I see Rdb or any other relational database being used on them.

        The reason being that VMS has a decent file system that means most of the time you don't need one, as RMS could do all that was needed.

        In fact it could do things many of the database systems (with the exception of DEC's Rdb) couldn't for many decades later; this was due to "clustering" not only pretty much being invented on VMS, but built into the core OS.

        I know that lots of the stuff I do today with crappy Windows/Linux boxes could be done easier, faster and a damned site more reliably on a VMS/AMD64 box.

        If only it had been done 10-15 years ago and sold for a few grand (same price as Windows Server OS's) there would be "fousands ov em".

  3. This post has been deleted by its author

  4. h3

    Transitive could have been the solution (IBM saw the risk and dismantled).

    Transitive had absolutely wonderful technology for running anything on anything else at almost native speed.

    They would have been a lifeline to either HP or Sun but IBM bought them (and basically dismantled them).

    (I guess IBM were worried about their mainframe business - They have the skill to do loads of stuff with virtualisation (Hell they have been doing it forever) that they choose not to so as to not harm their cash cow.)

    Apple used them for Rosetta

    I used them for a whole to run scientific apps that were only available for sparc after Sun stopped selling workstations at all.

    They are in a mess and only something really special is going to save them.

    (They won't take the risk).

    Itanium is a great arch for writing assembler for. If they managed to a quire a team of master chip designers I think they could use the itanium stuff as intermediate and then convert it to something better for real life performance. (Something like a better Alpha).

  5. John Riddoch

    Interesting parallel in some ways. I still think Sun's flip-flopping over x86 support cost it. I still think Sun could have saved their Unix biz by investing in a solid, scalable x86-64 platform with decent RAS features and plonking Solaris x86 onto it. Bringing some of their RAS/scalability expertise to Intel or AMD would help get the chips into a better state while drawing on the ever increasing clock speeds & compute power they brought but SPARC under-delivered on.

    They'd keep their SPARC business ticking over on life support for those that couldn't/wouldn't move until people migrated to Solaris on x86 which is still less of a wrench than moving to AIX, HP-UX or Linux on any platform.

    On the flip side, HP already had two hardware platforms out there (we still have a bunch of PA-RISC servers chugging away as well as some Itanium kit) and adding a third wouldn't go down well in some quarters.

    WRT the legal disputes, I'm waiting for the evidence to be revealed in the trial to find out what's really going on (or at least most of it) - at the moment, it's two spoilt children repeating "he said this" "he said that" "wah, wah, wah!"

  6. Anonymous Coward
    Anonymous Coward

    All a bit sad, innit?

    As far as I can tell, the story so far is that HP and Oracle swapped a kiss, not any bodily fluids, and Oracle is not under any contractual obligation to continue to support their DB on HP-UX/Itanium.

    Furthermore, it must have been painfully, painfully apparent to HP's management that Itanium was a dead end. This is one of those situations where having competition actually validates your own presence in the space: if Dell, HP and Sun were all selling Itanium systems, say, customers would assume that it was a reasonable ecosystem. Having HP alone selling Itanium makes it look, well, a bit sad.

    My advice (and admittedly, I'm not one of the MBA-holding corporate gods) is to ditch HP-UX, take all the goodies from it (and the other OSes, notably VMS) and create a better Linux than the other guy. Sort of what Oracle did with RedHat. I can imagine that an HP Linux with the sort of quality clustering that VMS had would be pretty attractive to some high-margin customers.

    1. Anonymous Coward
      Anonymous Coward

      Re: All a bit sad, innit?

      Agree, there is no contract, just a press release. Oracle doesn't sign anything that limits their ability to do what they want, when they want. All Oracle did was throw off HP's schedule for the end of Itanium and the OSs, HP had more of a 2014 announcement in mind and Oracle had more of a right now announcement in mind. Did Oracle do it to benefit Sun? No doubt, but it doesn't change the fact that everything Oracle said about HP BCS is true... regardless of motivation.

      They are attempting to bring the BCS software to x86 - Linux. HP can't develop their own Linux fork. They can do it technically, but they cannot entice the software providers to support HP Linux instead of Red Hat. Oracle gets a pass on Oracle Linux because they have the world's largest database business, so software application support was never a concern for Oracle. HP doesn't have much of a history in the Linux community either, certainty not like IBM or Oracle.

      1. Matt Bryant Silver badge
        FAIL

        Re: Re: All a bit sad, innit?

        And, news just in from Wunderburo's alternate reality - he knows nothing about Linux!

        "....HP can't develop their own Linux fork...." Anyone can make a fork if they want to. And hp Labs have more than enough knowledge seeing as they drove the Linux on Itanium effort.

        ".....HP doesn't have much of a history in the Linux community ...." So, apart from being the first vendor to offer Linux as an option to all their servers and their desktops, years before IBM, and apart from developing the version fo the kernel for Itanium, and apart from being the number one Linux on x86 b=vendor for so many years I've lost count, I suppose hp do have no presence in the Linux community. Oh, hold on a sec - that's masses more than IBM!

        Wunderburp talking out of his recturm again? This is my surprised face, honest.

        1. Jesper Frimann
          Pint

          Re: All a bit sad, innit?

          Come on Matt.

          HP involvement on Linux on Itanium is kind of non existent today.

          When I look in my bookmarks under Itanium, I find that

          http://www.itaniumsolutionsalliance.org/ gives me server not found.

          http://www.gelato.org/ has no activity on it..

          http://h20341.www2.hp.com/integrity/w1/en/os/linux-on-integrity-overview.html has an alarming amount of references to Linux on Proliant.

          So IMHO it's not like that expertise is there anymore.....

          And when it comes to the linux Kernel HP isn't even in the top 20 of contributors.

          Not that HP couldn't do a their own Linux distro. But they would have to build it up from the ground, unless they were to go and snatch up a company like Redhat, but personally I don't think that they want to shell out 10-15 Billions right now, for a company like that.

          // jesper

          1. Matt Bryant Silver badge
            Boffin

            Re: Re: All a bit sad, innit?

            ".....HP involvement on Linux on Itanium is kind of non existent today....." Unfortunately for Red Hat (and for Micro$haft), us customers had a preference for hp-ux on Integrity, especially on the enterprise servers like the Superdome. I'm told it was something like 90% of the Integrity servers went out of the factory with hp-ux rather than another OS, and that's pretty insane considering you could buy the servers OS-free and download and roll your own Linux if you wanted. From a commercial viewpoint, I think it was a case of "we're paying extra to get the best hardware, we might as well pay extra to get the best OS with the best support that goes with that hardware." I guess the same went for the developers, they just saw better dev support from hp and Intel rather than RH.

            As I've posted many times before, it wasn't a case of we COULDN'T do it on RHEL, it was a case of it being EASIER to go with hp-ux seeing as we had the option of buying tested stacks (like Amdocs with Oracle and Websphere on Serviceguard on 11i), with proven 24x7 support. We COULD do the same on RHEL (we kludged it toegther and tested it), and hp were happy to support the effort (I had two guys in hp Germany and a guy in hp Labs in the States to call on to get it working, plus a number of insultants from the local hp reseller), but we just felt more comfortable with hp-ux.

            As for Windows, the only area I saw that shone in on Itanium was large M$ SQL server implementations, and once hp started building large x64 servers like the DL785, RHEL and M$ saw the dominance of hp-ux in the Itanium market and decided to concentrate on the x64. For me, it was always a shame M$ never developed a fork of Exchange for Itanium, capable of really large mailbox numbers and ready to go head-to-head with Notes/Domino on p-Series, but I guess they saw that IBM were happilly trashing that threat to Exchange all by themselves.

            "....But they would have to build it up from the ground...." If hp ever wanted to get antagonistic with RH, they could just do an Oracle, clone the latest CentOS release, and add a few lines of comments, and - bingo! - hp-Lux. They could go a bit further and add some code to make it work better with other hp tools like Serviceguard (used to be supported with Linux and IMO was better than the standard RHEL or SuSE clustering). Thing is, hp are already the leading Linux server vendor, their tools like SIM and IC already integrate fine with RHEL (and VMware, and Windows), so they don't NEED to do an Oracle and clone a Linux.

          2. This post has been deleted by its author

        2. This post has been deleted by its author

        3. Anonymous Coward
          Anonymous Coward

          Re: All a bit sad, innit?

          "Anyone can make a fork if they want to. And hp Labs have more than enough knowledge seeing as they drove the Linux on Itanium effort."

          Read what I wrote and don't just respond selectively to parts that suit you. HP can absolutely develop a Linux fork, as I wrote. The question is: Which ISVs will support "HP Linux"? Certainly not Oracle, certainly not IBM, certainly not the broad base of ISVs, possibly SAP. It is not about technical capability, is it about getting every ISV under the sun to agree to port their software and support yet another Linux distro.

          "So, apart from being the first vendor to offer Linux as an option to all their servers and their desktops, years before IBM, and apart from developing the version fo the kernel for Itanium, etc, etc, etc"

          Yes, HP contributes to Linux in the same way Dell contributes to Linux, they put what other people have created on their servers. IBM has 1,000 developers actively contributing to the Linux community and basically created and donated major utilities like SystemTap (Dtrace alternative). HP has never done anything remotely similar. Saying HP contributes to Linux is like saying HP contributes to MS Server. HP makes boxes which run both OSs.... I am not saying they don't throw Red Hat on their servers, but that is not exactly substantial contribution to Linux as an OS.

          1. Matt Bryant Silver badge
            FAIL

            Re: Re: All a bit sad, innit?

            "....The question is: Which ISVs will support "HP Linux"?..." Now you're just being silly. If it has the common Linux kernel then anything compiled to run on the same kernel will be supported with the "hp-Lux". It's not like hp would have to re-invent the wheel, just add a few bits of their own. Do you seriously think that Oracle have to get a completely different binary for every app for their CentOS clone? Given hp's market clout, it would be trivial to herd up some devs to support an hp version of Linux. You really are displaying exactly why it is pointless trying to talk Linux with anyone from IBM.

            "....IBM has 1,000 developers actively contributing to the Linux community...." I doubt it. Please do link to the IBM page lsiting the numebr of IBM internals working on Linux projects. They may have a large number of devs feverishly writing code for Linux on mainframe, to try and keep the monopoly limping along, but I see SFA IBM effort anywhere else in Linux. Oh, did you mean on that tiny i-Series effort? What, that "1,000 developers" wrote a line of code each? LMAO.

            ".....HP has never done anything remotely similar...." Apart from the Tru64 Advanced File System you mean? Which is a much bigger contribution than a sad clone of a tool most Linux users would simply not require.

    2. Anonymous Coward
      Anonymous Coward

      Re: All a bit sad, innit?

      They probably could do that with HP-UX, but with stuff like VMS and NonStop which are quite unique, I wonder to what extent they could just lift the desirable features out of them and dump them into Linux. I'm pretty sure something like VMS clustering would not be trivial to implement in Linux.

      1. Anonymous Coward
        Anonymous Coward

        @AC 13:43

        "I'm pretty sure something like VMS clustering would not be trivial to implement in Linux."

        Everything Linux/*nix boxes can do, VMS can do using little more than its C run-time library.

        You would have to change just about everything in any *nix OS to do half of what VMS can do out of the box, they are just that different. Even the simple things like Logical names (decoded at system service level not the shell), enforced file and record locking (cluster wide), indexed files (a bit like C-ISAM) with all the OS utilities to create/tune/access them from any language.

  7. Anonymous Coward
    Anonymous Coward

    Wow...

    There is some pretty damning stuff is this trove of e-mails:

    Donatelli at HP – "the regions are unaware of the situation with Itanium and the impending end of life. The Itanium situation is one of our most closely guarded secrets"

    This is the gem, Martin Fink at HP to Skaugen at Intel:

    Kirk-

    Call me when you land. We've been told that we can't use the following line.

    "Intel added that it at no time communicated to Oracle a change in commitment to the future of the ltanium processor family."

    This is a MAJOR, MAJOR issue. We need to be able to tell the market that you never told Oracle about EOL plans for Itanium. This is a CRITICAL element of the HP/Intel relationship. I don't view this as optional.

    Martin

    Martin Fink I Senior Vice-President & GM I HP Business Critical Systems I 970-898-7076 I martln.finI«8thcom

    Executive Assistant I Ingrid Busch I 970-898-0782 I ingrid.busch6hp.com

    and this from Scott Stallard at HP to Fink at HP:

    1. call it Poulson, whatever it is, paint a good picture on why me made change made it earlier for HP's roadmap or whatever

    2. don't possibly signal to world end of PF roadmap. or you kill tw success and the bcs business

    3. need a placeholder for kitson, since you went and announced that without our knowledge keep a cpu team around and keep them busy otherwise you are pulling out the full tw and poulson teams within 18 months not a soft landing, a crash landing!

    thanks

    SS

  8. asdf
    FAIL

    fail.

    Is the HP way now heavy turnover in the CSuite followed by massive layoffs every new CEO? Tha'ts the kind of the thing you want to see in your mission critical systems vendor and their bottom of the barrel India support. How far the mighty has fallen.

  9. This post has been deleted by its author

  10. Anonymous Coward
    Anonymous Coward

    Back at you Oracle

    Now HP has released a bunch of internal Oracle docs. Basically executives talking about legacy Sun being dead. Legacy Sun "baaloooowws"

    http://www.scribd.com/doc/93811611/HP-Itanium-docs-pdf

    1. Matt Bryant Silver badge
      Pirate

      Re: Back at you Oracle

      So that's Keith Block, Oracle's Executive Vice President, North America Sales and Consulting, admitting that the Oracle move has no technical merit, that the intention was just to "fuck hp". Love to see how Larry spins that one!

      1. Anonymous Coward
        Anonymous Coward

        Re: Back at you Oracle

        Yes, it will be difficult to explain why two of their executives are writing about how much Sun blows, it is "dead dead dead" and it is a "pig with lipstick... at best."

        Still, thinking Sun blows does not necessarily mean that they pulled Itanium support as a result. It is possible to think Sun blows and HP Itanium also has no future... that is what most people think. Sun being bad doesn't make Itanium good.

        1. Matt Bryant Silver badge
          Boffin

          Re: Re: Back at you Oracle

          "....thinking Sun blows does not necessarily mean that they pulled Itanium support as a result...." Ah, but you forgot that Larry has always claimed this was about helping their customers and not about shafting hp, whereas hp has always said it was exactly the latter. It looks very hard for Larry to keep on pretending this wasn't an attempt to use their database near-monopoly to try and pinch server business from hp.

  11. P. Lee
    Terminator

    Make it better

    All the commercial guff hides the fact that you need to do things better in order to compete.

    Regardless of the past, HP need to look to the future. If HPUX is a better enterprise system than others then leverage it, don't spend time wailing about what might have been.

    Build some reliable hardware, stick HPUX on it. If you can do a better job with HPUX than with Solaris or BSD or Linux or Windows, then do so. Stop spending billions on lawyers and put it into development to make your own products / services better. If you want a stick to beat Oracle with, put development time into postgres.

    Even if the courts made Oracle do what HP want, Oracle obviously will do the bare minimum, the relationship is shot and it won't end well for HP.

    1. Adam White

      Re: Make it better

      I think the point here is that platforms are only of value if they run applications people want to buy. HP can't expect to "build it and they will come" now that Oracle has shown it can and will refuse to support them.

      1. Anonymous Coward
        Anonymous Coward

        Re: Make it better

        Yeah, but people can't blame Oracle for the downfall of Itanium. They might have expedited the process, but it was going out now or a couple of years from now. HP couldn't even get their BFFs at Microsoft to continue supporting Itanium. Itanium is not competitive today, so the situation is more: HP built it, the ISVs came, HP let it get uncompetitive, the ISVs left.

        HP has this product end of life strategy they call "store next door" which is what they did with PA-RISC to Itanium, what they are currently are doing with XP and EVA to 3PAR and what they wanted to do with Itanium to x86. The idea is that they sell the old, dead product in parallel with the new product for a couple of years to try to get voluntary migrations. As in "We will sell you PA-RISC, if that is what you want, but have you seen the wave of the future Itanium chip... "We will sell you XP or EVA, if that is what you want, but have you seen 3PAR." After enough people have moved, they announce EOL. As the e-mails demonstrate, they even used "store next door" by name, that is what HP planned to do with Itanium and the legacy OSs but they thought they could hold on to them for a few more years before bringing out the x86 BCS gear. Now that Oracle has pointed out what they are up to, they are left with a high end offering everyone knows is dead and a replacement that does not yet exist.

  12. Anonymous Coward
    Anonymous Coward

    Tru64

    Just port Tru64 to x86-64 and I'll be happy! Or some source code!

    1. Anonymous Coward
      Anonymous Coward

      Re: Tru64

      HP will do you one better, they will sell you all of the Tru64 IP at a very reasonable price.

  13. SplitBrain

    Matt B...

    Even you can't bullshit yourself out of this one...

    SP/L or whatever it is...

    1. Matt Bryant Silver badge
      Happy

      Re: Matt B...

      Aw, one of the ickle Sunshiners has gotten over the shock of the Sunset enough to come and post! No surprise to see PeeBrain still maintaining the Sunshiner tradition of adding nothing to the conversation.

      Guess I'll have to remind him of the failed Sun port to Itanium (it happened), the failed efforts of Sun to convince the market to buy Slowaris x86 instead of RHEL or SLES, and the way Sun came begging Carly to support Slowaris x86 on ProLiant (beacuse Sun's comedic efforts at making x86 servers of their own were just too funny for words!!).

      Fact is, even should hp fold hp-ux or Intel can Itanium, both will still carry on as IT industry juggernauts as both (especially hp) are diversified, whereas Sun shrivelled and died because they had a stupid fixation with SPARC. ROFLMAO!

      /SP&L

      1. fch

        Re: Matt B...

        As an ex-Sun(shiner) I take offense at the statement that the Solaris/Itanium port _failed_.

        It ran just fine in the lab. Noone wanted to have it, not a single then-Sun and then-prospective-Sun customer asked for it (very different from "canning" Solaris/x86 a few years later).

        Sun simply decided that there'd be nothing to sell, so it never was pricelisted / made available.

  14. OzBob
    Unhappy

    OS as a commodity now

    I work in HP-UX (have done since 95) and while they have improved it in the latest version (and done some clever things with virtualisation), they have not really sold themselves as the "big iron" provider. Redhat and other Linux-based OS's are whittling away at the smaller databases and I believe some of the larger NHS systems are HP-UX, but even my employer is looking to port away from Itanium to more generic Redhat and Wintel (tech support issues and supply problems with HP contributing to the decision).

    The OS seems to be a commodity now to run applications on, rather than a specialised item, and HP are not adding enough value to justify keeping their OS. if they could add that value (specialised tuning tools and consultancy, cheaper hardware, better support, integration with other packages) they could retain their customer base.

    Right now, the HP engineers I talk to are demoralised and unmotivated, and the Software Support Staff have an accent I could cut with a knife (I now claim the phone line is bad and could they please email me their response). Bill and Dave's company as they knew it is dead.

    1. Anonymous Coward
      Anonymous Coward

      Re: OS as a commodity now

      For smaller DBs and whatnot, no question, they should be on Linux. For mega-DBs, like NHS, and other mission critical workloads, there is still a place for Unix. The Unix tool sets and RAS hardware/software features still have an edge on Linux, albeit not as large as it was a few years ago. Unix systems still scale way beyond Linux - x86 unless you have some really boss clustering tech, but that also adds complexity. Unix support is generally still better than Linux support. The biggest issue is that Linux, on x86, is still a roll your own. You have to be responsible for building parts of the OS that used to be included. It not only adds extra integration, but the industry spats are issues. Oracle demands Oracle Linux and deeply incentivizes OVM, VMware and Oracle hate each other, Red Hat and Oracle hate each other, Microsoft and VMware are parting ways, etc. It is hard to know who is going to be the winner of all of those fights in a few years. It might just get really fragmented.

  15. Ramazan

    demoralized

    True, and not only engineers are demoralized. HP managed to fuck up with some formal procedures after winning a tender here. So we're looking for a better contractor at the moment to do the job.

  16. IGnatius T Foobar
    Linux

    HPUX is a legacy OS

    HPUX is a legacy operating system. There would be no uptake of an x86 port at this stage of the game because any ISV's that are porting to x86, will simply port to Linux. The bottom line is that Linux won the unix-on-x86 contest a long time ago.

    It's the 21st century, people. No one makes money selling operating systems anymore except Microsoft, and even that is on its way out. There's plenty of money to be had in supporting those operating systems, as the likes of Red Hat and even IBM have discovered.

    Of course, we all know that HP lost its way a long time ago.

This topic is closed for new posts.

Other stories you might like