Re: Looks like some big decisions have been taken.
HP will forge Integrity and Superdome machines based on Intel's Xeon processors in addition to the Itanium machines it currently is selling, and plans to upgrade through at least two more processor generations. Not that we didn't warn you. El Reg anticipated this when the "Tukwila" Itanium 9300-based Integrity and Superdome 2 …
HP has decreased its R&D from 6% to 2% in recent years and cannot afford nor do they have they staff to keep up so they are moving their Integrity team to work on x86 and their HP-UX team to get their middleware certified on Linux. They will concede the O/S to redhat but want to keep openview and serviceguard.
The next two itanium chips will go into the current system but dont expect an upgrade to the SX3000 chipset. All those folks will be working on the x86 glue chip.
The "enhancements" needed in xeon are for the non-stop not HP-UX. That will be obviously ported asap. VMS has obviously finally found its end date just like Tru-64.
Just like the old wendy's commercial said "wheres the beef" or the clinton campaign slogan "it's the economy stupid" this time "it's the applications" and HP-UX is the dog that don't hunt and wont hunt on x86 either. Oracle has obviously said sorry HP but even if you port HP-UX to Xeon we still won't support it.
Funny how Oracle basically replaced HP-UX support with z/OS support this year. After all the mainframe migration claims its the mainframe that ended up replacing HP-UX for Oracle.
Oracle wants to be the proprietary old IBM
HP finds itself being the nearly bankrupt IBM that people wanted to break up.
and, IBM finds themselves doing everything right with the highest market cap ever.
If you think about it the only real "engineered solutions" left are Mainframes and Power systems
This HP announcement is basically the same as Intel's move in April
Intel Relocates Itanium Engineers to Xeon Projects .
Mission-Critical Server Specialists Now Work on Intel Xeon
04/17/2011 "Nearly all the Itanium engineers, save a small development team working on Poulson and then rotating over to Kittson, have been redeployed on Xeon-related projects,"
Oh dear, Alli, have you been drinking too much rum down by the poolside?
"HP has decreased its R&D from 6% to 2% in recent years....." Really? And how much of that relates to Itanium, Integrity or hp-ux? You can't say. In fact, seeing as your "recent years" fudge is so wide, I'm betting you're comparing over the period when hp had Itanium development in-house (along with the Comapq Alpha-related products) and then comparing to after hp shipped the Itanium/Alpha teams off to Intel. How much of that research budget is for ink products, phones, tablets, PCs, memristor tech? You don't know and wouldn't want to say, because that would poke a big hole in your FUD. It's about as silly a line as me claiming Power is under threat because IBM have been forced to slash their computer lines in half in recent years. Sounds bad for mainframes and Pee series, no? IBM have, but only because they sold the PC and low-end server bizz to Lenovo. A little context goes a long way to poking holes in non-arguments.
"....The next two itanium chips will go into the current system but dont expect an upgrade to the SX3000 chipset...." Maybe that's because the SX3000 chipset was designed to scale across the next two generations from the start, so doesn't need an upgrade. Unlike Power chipsets, which had to be changed for the jump from Pee4 to Pee5, from Pee5 to Pee6, from Pee6 to Pee7, and from Pee7 will have to change AGAIN for Pee8. Maybe IBM should talk to hp about designing a bit more upgradability into their servers? Or maybe IBM just banks on ripping off their customers with every upgrade. After all, every time you upgrade AIX you have to go to the latest Pee-series CPU to get the full benefits, so if IBM are forcing that kind of upgrade burden on their customers I suppose charging them for new motherboards (sorry, "books") isn't much to add to the pain.
"....The "enhancements" needed in xeon are for the non-stop not HP-UX. That will be obviously ported asap. VMS has obviously finally found its end date just like Tru-64....." Love the evidence you give for those wild presumptions. Oh, actually there's no evidence at all! What a surprise - not! Currently, there's a longer public roadmap for all three of hp's Integrity OSs (hp-ux, Nonstop and OpenVMS) than for AIX, and a longer public roadmap for Itanium than Power. Gee, that must mean IBM is getting out of enterprise systems, right?
"...."Nearly all the Itanium engineers, save a small development team working on Poulson and then rotating over to Kittson, have been redeployed on Xeon-related projects,"...." Not surprising, Poulson is taped out and ready to roll, so the development staff have been moved to new projects until the main development phase on Poulson kicks off. By the way, how close to being taped out is Pee8? Oh, it isn't. Best IBM will publicly admit is it is "under development" for release at "an undisclosed date". Ooh, concrete stuff, I can really plan on that then!
Poor Alli, you're obviously suffering from a serious combined case of overwork and sunstroke! Looks like you should have taken a holiday a long time ago.
Taking into account that biggest customer for VMS is US Military and Navy I would be very very surprised if I would see end of VMS life anytime soon.
It does look like IBM is well on it's way back to being "The Computer Company". HP and Oracle are both engaged in long term self-destruction by pursuit of short term result.
HP is especially hamstrung by their dependence on Intel.
If you want to race Formula 1, you have to build custom cars. So long as HP have to convolute their OS to fit Intel's processors, they will be at a disadvantage to IBM in the Enterprise Mainframe market. And regardless of what happens at the user interface, he who rules the systems of record, rules the world.
".....HP is especially hamstrung by their dependence on Intel....." Simple debunking question for that load of claptrap - where is the IBM x64 design with the scale of the coming hp server using the Superdome2 tech? Will any IBM x64 server offering have the ability to use not only software partitioning but also hardware partitioning? Will any IBM x64 server also ofer the ability to mix Power with x64 hardware partitions? Looks like IBM simply won't turn up to that race, maybe they'll ask Lenovo to do it for them. Not surprising, really, as such a box would seriously threaten their golden egg mainframe business, where IBM exploit their customers to prop up their P-series bizz.
I would say that this is because they are not confident of winning the spat with Oracle.
"They want HP to make and sell Itanium-based machines running HP-UX, and ditto for those who sell OpenVMS and NonStop apps."
What a crock of shit - an excuse for not doing it. Why would you want to have teams that need skills for two UNIX's, especially HPUX which has far less supported apps than AIX and Solaris. Yes it does before the trolls start claiming it is fully supported (and I'm not even thinking of Oracle).
Similarities to Symbian, in that it got quite good in it's old age but it was too bloody late and missed the boat.
Hmmm - I Can't reply to or down vote the IBM Shill (Allison) post - I suspect a conspiracy! ;o)
Allison - can you produce a written statement from either IBM or Oracle promising to continue to produce Oracle software on AIX?
Those in glass houses etc...
Just the 2 more gens of Itanium to see them most of the way into this decade and no more HP-UX and Itanium, it's all been expected for some time, although I suspect a certain HP fanboy will roll through shortly, frothing at the mouth in denial as always proclaiming Itanium and HP-UX shall continue on forever. Or as usual, when it's a negative HP story he will stay mostly schtum..
However, a big boy x86 box with 32 sockets+ is very, very tempting, and if HP get's there first with a good commercial offering then they will sell shitloads of them. Red Hat on one of these beasties would be awesome just for pure grunt or large scale consolidation.
HP will continue to be the worlds Premier Tin Pusher as it is now. But alas, no OS, no Database, No Middleware, no Apps of any real use or any market penetration at all mean that tin pushing is all HP will ever do. Their value has plumetted by over half over the last year and it ain't gonna get any bigger now. Not a hell of a lot of money to be made in pushing tin.
see IBM/Oracle for a lesson in how to make some money in this game.
Was that my cue? Sorry, I was waiting for something worth responding to, but I was falling asleep reading the usual FUD. Anyway, on with the fun!
"Just the 2 more gens of Itanium...." Itanium has Poulsen taped out, Kittson in development. IBM has one more gen of Power "in development", Pee8, which doesn't even have a public release date. Snoreacle are still trying to make T-series actually perform the way us customers want, i.e. with some single-threaded performance, and their roadmap was too late and - frankly - too unbelievable. Sounds like hp-ux and Itanium products have a much sounder future than their competitors.
".....when it's a negative HP story....." LOL! It's a TPM article, he can't write anything about hp without it somehow meaning the death of Itanium. "Meg Whitman went to the bathroom today? Must have been because Itanium is going down the pan...." etc, etc, etc. Please try finding a single one of his articles not written with Big Blue blinkers firmly in-place. The funniest ones are where he paints up slippage on IBM products as "planning" (such as http://www.channelregister.co.uk/2011/10/11/ibm_power_systems_enhancements/). The actual bones of this article is that hp are planning on using the same Superdome2 tech for Xeon, but somehow that massive market advantage (where is the IBM 32-socket Xeon server, or the Snoreacle one?) becomes "Itanium is dead, hp-ux is going to be killed". It's not just a rediculous jump, it's about as far fetched as asking the head of Xeon development at Intel if he will be spending any time on Itanium and then taking his negative answer as an excuse to stop supporting future development of your software on that platform. Now, who would be silly enough to do that?
"....no OS, no Database, No Middleware, no Apps...." Strange, but hp seems to make a lot of money deploying OTHER companies's databases, middleware and apps. In fact, more IBM software is sold by hp for use with hp servers than by IBM itself. Vendor with the largest share of the servers used for Linux? Hp. Ditto for MS Server, SQL and Exchange; Oracle DB and RAC; and SAP. Maybe it's because hp offer the customer choice, whereas IBM (and now Snoreacle) are all about lock-in. Not that hp don't sell their own software, after all the recent hp Software results for 2011included a 28% growth with 27.7% margin in what has been a tough year. I think you'll find that IBM's report for Q3 showed only 13% growth for their software bizz, so hp is growing theirs at more than twice the rate of IBM. Even Oracle only managed 23% software growth in their best quarter this year.
"....Not a hell of a lot of money to be made in pushing tin..." Good thing then that hp has tech like memristors coming down the line. Not to mention that printer and ink bizz that winds up the Sunshiners so much.
That'd be a definition of "homegrown" that I was not previously aware of.
At $500m revenues per quarter the installed base of customers has to be less than about 1000 in the world because an Itanium cluster is going to cost at least a few million bucks on the average. So we're not talking about folks we know or care about, likely.
If you have a problem the Itanium is a solution for, you already have a vendor to sell you it, so I'm not going to waste my day looking for you.
HP Unix it will be a loss, but that war's been fought and Linux won - oddly enough through SCO. The provenance of Unix is now so tangled that the knot cannot be undone. But that's OK. We still have a way forward without reinventing everything.
$500M x 4 Quarters = $2000M, divided by 1000 customers - $2M per customer
Wow, deep statistical analysis there!
So every Itanium customer replaces their kit every year do they?
An Itanium "cluster" costs "a few million bucks" does it?
With that level of inisght you should go work for Gartner or IDC!
Hint...the smallest Integrity server can be had for as little as about £13K, only fully loaded Superdomes with 128 cores and 2TB of memory, plus _all_ the HP-UX software get up to around the $2M mark, and thats without discounts
Not arguuing with your other points, but I suspect HP has a _lot_ more than 1000 Itanium customers
Actually, pricing from hp can be fun. The larger the system and the deal, the bigger the discount you can hammer out of hp. As a comparison of performance-per-quid, the smaller servers actually come out worse as they are less steeply discounted (unless you buy several hundred in a go). It's pretty much the same for all the vendors. At least that holds until you get to discussing support costs, as large servers usually go into business critical roles and so you usually add 24x7 support to them, whereas smaller servers may go into departmental roles and only be on cheaper 9x5 support.
But, as a guide to market penetration, hp claim to have Integrity in 70 of the current Fortune 100 companies, which is more than Snoreacle can claim for T-series, and more than IBM can claim for Pee-series.
Sorry, been busy. Some of us do have to work for a living, you know! Well, every now and again at least.
Everybody looses if HP quits HP-UX, because competition will be less. This is actually not good. I would not like a scenario where HP-UX and IBM AIX is dead, because then there would only be Solaris left. And Oracle would have a monopoly. No one, except that single monopoly vendor, benefits from a monopoly.
Competition is healthy, and forces the vendors to be better and cheaper. And us customers benefit. Now IBM POWER7 gear is cheaper than POWER6 gear was, because IBM had to face hard competition from Xeon and SPARC. The same with Intel Sandybridge CPUs, they are cheap because of competition from AMD Bulldozer. However, Bulldozer turned out to be not that good. Thus, Intel lost their fear, and I would not be surprised if IvyBridge has a higher price than SandyBridge had.
There is an interesting conspiracy theory I read on the internet, and everything that is written on the internet is true, right? ;o)
Both IBM and Oracle seem to attack HP. And Apple and Oracle is attacking Google Android. It is almost as several big companies cooperate, and then attack one single company. For instance, HP has lost customers. In my big finance company, HP hardware is now forbidden to buy, HP has lost us as customer. Just as Sun hardware was forbidden to buy. They are black listed. Oracle is white listed. Who makes this list? You sure dont want to be black listed.
Let us assume there is a small and powerful network that controls other companies. If this network in the year 1980 decided to bet on a small company, say Microsoft, and help it to become big, then MS would thrive. Because MS would get lots of customers. If the network, later switches interest to Apple, then Apple would become big. Or Facebook. Or Google. Or Linux but the network did not like FreeBSD. And the network has of course invested and bought stock. Now assume the network looses interest in HP, what happens? IBM and Oracle gets a "go" to attack HP. The network shorts stock in HP. HP looses lots of customers, because the network say "HP is black listed".
Silly theory, isn't it?
Here is a new PhD thesis. He has researched for several years, and found out the following. He scanned huge databases with economic information and noticed which companies owned who, using mathematical tools. After scanning 37 million companies, he could see a pattern. It was always the same 147 companies that owned other companies. Those 147 companies were mostly investment banks on Wall Street. Yup, you recognize them: Goldman Sachs, JP Morgan, etc. This network also has tight and close cooperation, he showed this too. There are numerous cases when Goldman Sachs and JP Morgan and other banks, for instance, shorts commodities such as Silver, etc etc. Simultaneously. What is by coincidence, or was it cooperation?
Here is the fresh research, including the PhD thesis, done by some complexity theorist researchers:
An oligopoly is a market form in which a market or industry is dominated by a small number of sellers (oligopolists). The word is derived, by analogy with "monopoly", from the Greek ὀλίγοι (oligoi) "few" + πόλειν (pólein) "to sell". Because there are few sellers, each oligopolist is likely to be aware of the actions of the others. The decisions of one firm influence, and are influenced by, the decisions of other firms. Strategic planning by oligopolists needs to take into account the likely responses of the other market participants
I did not understand your post. What is your point about oligopoly? Is it good? Is it good that HP is attacked by several companies?
"HP is vulnerable" said Larry. He also said "I dont want to taunt IBM too much, they have great products". It almost is as if IBM and Oracle has agreed on cooperation and attack HP together. This is silly, the world is big enough for HP, Oracle and IBM.
I thought you were in Jamaica feasting on tender lamb ?
"I thought you were in Jamaica feasting on tender lamb ?" Alli probably only took IBM Red Books as reading material, it's the only thing that would make me bored enough to be posting from Jamaica!
Alli, if you're bored and in Kingston, I'd recommend the Quad Club in Trinidad Terrace. Four floors with different music on each, even '70s stuff!