Oracle's standing in the EMEA server space took another battering during Q1 as sales collapsed in a market that posted double-digit growth, underpinned by high-end and mid-range shipments supporting virtualisation and cloud deployments. The US behemoth saw revenues plummet by almost 20 per cent to $241m, pushing down its market …
Since the Oracle takeover Sun have become even more difficult to deal with, it takes them days to turn around quotations for anything out of the ordinary, and their maintenance is ridiculous.
You might think every order has to go back to the States to be authorised, every time you make even the smallest change.
Compare and contrast our IBM experience whose pricing is aggressive, and response immediate.
Pity really great kit, crap sales process, thanks Larry.
I don't work in the Sun/SPARC area, but I have a friend who does at a US
government agency. As a result of Oracle's acquisition, they are re-evaluating their buying practices because they can't afford the new prices, and even if they did, Oracle make it impossible to order.
waiting for the other shoe
""The whole RISC and UNIX market remains in the doldrums with customers continuing to adopt a wait and see attitude [as they assess recent tech advances on x86, at chip and systems level, designed to boost reliability, availability and security]," [Nathaniel Martinez] told El Reg.
Nothing to do with Sun's customers having the jitters that Oracle might ditch Solaris next? honestly?
Oracle ditching Solaris ? A toast to that !
... them finally taking their anti-midas touch off it.
Outsourcing Solaris development and support would at least give x86 customers a chance to deploy it without that sword of damocles above them - support cost hikes for continuously degrading service.
It's Friday, after all !
Hate to point this out, but the phrase "on their uppers" generally means "broke", deriving from the idea that one is so poor that the soles of one's shoes have completely worn away, leaving one "on the uppers". Growth of between 10 to 33 per cent doesn't really seem consistent with that?
May be down to
the price is too high
the lead times too long
the ordering process a joke
the support costs ridiculous
their attitude to customer stinks
partners are not selling due to all the above
AND THEY SEEM NOT TO GIVE A STUFF WHEN THIS IS POINTED OUT ON A DAILY BASIS
And the future of Sparc is?
Oh dear, and so predictable. Might we expect that Oracle will now decide that SPARC, along with Itanium will not be supported on new versions of their database and middleware software...
Well, maybe not, but it rather goes to show that if Oracle were seeking to gain market share for SPARC from HP-UX Itanium machines by their announcements, then all they were likely to do was to accelerate the migration of customers to x86/x64 platforms. Since SPARC servers (especially on the M series) are now very much premium-priced products it's not going to look like an attractive alternative. Of course it's a real pain to move large, enterprise databases from one hardware architecture to another, but if it has to be done, logic says go for the one processor architecture which is pretty well guaranteed to be around for whatever counts as the long term.
In the meantime expect Oracle to milk the SPARC market as a legacy one, much in the same way that IBM do with mainframes. There will always be a significant number of customers that can't or won't migrate for cost or risk reasons.
Speaking as a (very) ex-sun employee...
...who quit the company quite a while ago, years before this acquisition was even a twinkle in Ellison's eye, so to speak, I know I find myself in the rather sad position of having to recommend people NOT purchase "Sun" equipment anymore. I loved Solaris, and still feel that it is a more polished, professional, and generally clean OS than Linux. I know from personal experience that once upon a time Sun servers were some of the most reliable non-redundant computers you could find, and when they did break, Sun service was just Top Notch. Unfortunately, that just hasn't been the case for some time, and ever since Oracle purchased Sun, I personally have heard a steady increase in complaints that match the ones on this thread.
It really is sad; I'd like to think that Oracle is going to keep something other than just Solaris-for-Oracle-DB out of their Sun acquisition. On the other hand, I'm not certain that it really fits their business model to do anything else. Maybe trying to eek some money out of Java. :(
I still support those lovely boxes you helped create. OK a few things to learn (I like that as it's a new challenge) but they are nice machines. And once you can sing with them the power is at your fingertips :-).
'Tis truly sad that things are going this way.
The x86 boxes that will replace the Sparcs will undoubtedly result in out of hours downtime. Guess who'll be on the receiving end of it ? Less time in the pub is likely. Hence pint icon in protest.
So, Larry give us back our vital Sparcs !
No surprise here...
They're doing it to themselves I think.
My "IT roots" so to speak are fully tied into Sun, their Solaris OS to be precise. In fact; a Solaris course which I was sent to years ago was also the main thing which really sparked my interest in Linux (again). And I've been using Solaris for a long time now, ever since '93 - '94 I think, both privately as well as professionally.
I loved it so much that when Solaris 10 was fully out for both Sparc and PC ('x86') it didn't take me long to embrace it. Sure, after the launch I was left a bit disappointed about zones and linux-zones not being available at that time, but Sun sure managed to make up for all that in my opinion. MORE than that even (think ZFS for example). And although Solaris 10 could be used free of charge I still applied for a (basic) support contract. Not that I needed hand holding; but the continues updates were sure a very good way to keep the system up to date all the time (Sun did provide free updates too, but with a support contract you didn't have to wait for the updates to become "bundled and distributed", instead you could update the moment you felt like it).
Oracle's prices for this very same subscription are almost threefold of what it was, and I get a whole lot less service in return. And as much as I love and admire Solaris (also their ties into Java has always impressed me) I hate ('dislike') Oracle in almost the same amounts (maybe even a little more).
My move was simple.. I moved all my (2) servers over to another OS and company-wise (self employed) we also dropped advice and support for Solaris alltogether. Its just too much of a risk now; that and the fact that I don't trust Oracle for one cent.
Servers? Right now we're focusing ourselves on both Windows and BSD servers. There are places where a Windows server can excel just like there are places where BSD would be much more preferred.
But no more Solaris for us. And given the way Oracle treats its customers we want to have as little to do with Oracle as possible. A shame about Java but I can't (nor do I want to) drop that either, but if there comes an opportunity of some sort....
My stance? Oracle only thinks "big" without realizing that sometimes big can start small. Their loss.
CISC RISC EPIC processors?
What are they then? All things to all men? Or did someone leave out a comma and an "or"? And what CISC architectures are they talking about anyway, if not x86/x64?
SPARC is toast and you know it.
SPARC is toast, everybody knows it. It's exa-data this exa-larry that. SPARC will be sold at huge premiums for customers that are too afraid or in IT parlance, "risk averse" to migrate aware from Solaris at a minimum or Oracle DB for reals...
Jeez, I think Oracle is actually hated more than MickeySoft, who'd a thunk it ?
They don't understand their customers
We run sparc for one reason and that is its hardware stack mess up many of the buffer overflow exploits. We didn't care that it was slower, and we didn't care that it cost more. We did like the rock solid hardware and the ability to strip Solaris down to a very, very tiny set of files that would run our applications. Then came Solaris 10 with all the new features and the constant patching and the constant breakage. We can't run Sol 9 on their new hardware so we aren't buying any new hardware. We can cope with the stupidity of Oracle software only support but I need hardware and software that doesn't have to be rebooted every few weeks to maintain its own security. From now on its BSD on telco grade sparcs as long as we can. Then we will most likely go with Power or something else a bit less mainstream.
'Then we will most likely go with Power or something else'
Chuckle, think you're having fun on Solaris/SPARC? Wait until you've experienced the sheer misery of AIX/Power - nothing works properly, ever. You patch for issue A, you patch for issue B, issue A crops up again. The hardware is little better either, I swear IBM factory workers wear nylon clothes and shuffle their feet a lot when putting the things together - its the only explanation for the atrocious failure rate of components that we suffer with.
I agree with anonymous that Power went through some bad times with AIX 6.1.3, but that was mid-2009 and things have improved (I know that not-everyone went through it during that time, but the 6.1.5 and 6.1.6 have been spotless). IBM seems to have the most difficulty during times of intense hardware changes.
Solaris 'homeys' will have difficulty leaving the mother ship, but Oracle is struggling with fixing the problems that happened during the reign of 'pony tail' .
IBM is prepared with higher GHz CPUs, higher thread-count processors (96/socket by Fall) and fancy I/O for Power 7+. IBM has an advantage in that they basically monopolize the fabrication of systems from Power 7. They can over-engineer the boards to get performance advantages.
They will NOT be dropping the price dramatically, though. IBM's investors were clearly surprised by the price-drop in early 2010 and have exerted pressure to not-have IBM attempt to make Power 7 a dollar-to-dollar competitor to Intel's commodity business. They want Power to sell at a substantial premium over x64.
I do not agree with much you say, some things are just plain wrong.
"...IBM is prepared with higher GHz CPUs, higher thread-count processors (96/socket by Fall) and fancy I/O for Power 7+. ..."
Dont you know that GHz does not necessarily imply good performance? Do you remember Intel Pentium4 which had high GHz, and very low performance?
Or, do you happen to know about the new IBM z196 Mainframe cpu "the worlds fastest cpu" at 5.26GHz with 300MB cache? It has lower performance than a Intel Nehalem-EX and costs much more.
Have you heard about the slow POWER6 at 5GHz? You needed six POWER6 servers with 14 POWER6 cpus at very high clock, to match one Sun T5440 with four 1.4GHz Niagara T2+ when we talk about Siebel 8.0 benchmarks.
Or, the IBM CELL. You need thirteen CELL cpus at 3.2GHz to match one Niagara T2+ at 1.4GHz in string pattern matching. That is, 42GHz worth of aggregate GHz to match 1.4GHz. That is a factor of 29.7. You need ~30 times more GHz if you use IBM cpu CELL to match one Niagara T2+. that is quite bad.
So you are wrong when you imply that high GHz equals good performance. IBM gear disproves your claim.
Regarding IBM having higher thread count processors (96/socket) - what do you mean? I mean, Niagara T3 has 128 threads per cpu. To me it seems that 128 threads are more than 96 threads. Maybe you dont know about Oracle cpus? Niagara T3 has several world records today, beating POWER7 in some benches.
"....[IBM] will NOT be dropping the price dramatically, though. IBM's investors were clearly surprised by the price-drop in early 2010 and have exerted pressure to not-have IBM attempt to make Power 7 a dollar-to-dollar competitor to Intel's commodity business. They want Power to sell at a substantial premium over x64...."
What are you talking about? POWER6 servers costed like 5-10x more than x86 servers. Now the POWER7 costs 3x more than x86 servers. Today the Intel Westmere-EX 32nm is something ~10% slower than POWER7:
The 22nm version will be 40% faster according to Intel. In that case, it will be faster than POWER7. IBM has no choice but keep lowering their price. Why would anyone buy a slower and 3x more expensive POWER7 server than a cheap faster x86 server?
In a couple of years, Intel and AMD 20core cpus, will easily be faster than everything IBM has. That is when IBM will see their POWER sales dwindle, and they start to lower the price. Finally, x86 will be very much faster and cheaper - that is the point where IBM will kill POWER and AIX. As IBM executives officially said: they will kill off AIX.
Then there is no reason to continue development of POWER. It will be killed off too. IBM must continue lower prices in the future. Gone are the high margin AIX systems. Left are the slow IBM Mainframes which has a high margin.
As I said, I dont agree with you much. Much of what you say is errorneous.
You keep repeating the same and same things Keb. Most of what is totally irrelevant to what Beachrider stated in his post.
And with regards to comparing Westmere-EX to POWER7 then sure that is a good and valid comparison, as they are both current processors. Although POWER7 is soon to be replaced by POWER7+.
But when that is said.. then Power7 is 15% faster on SAP SD 2 tier at 4 sockets, and it's a whopping 47% faster at 8 sockets. And at 16 and 32 sockets.. well there Westmere-EX isn't really a competing products now is it ?
And then you pull out Next generation of 22nm Xeon and say that it'll be faster than POWER7.. well to be Quite honest if an Xeon processor that is going to be shipped in servers in what.. 2013 isn't going able to match a POWER processor that shipped in early 2010.. then well.. I am sorry.. then it's not a very competitive product.
Just Ditched Sun/Oracle for reasons cited
Moving to RHEL on VMware for development teams. Oracle (are you people living on the same planet?) took 60 days to reply to a simple request for support on existing kit. Management was not amused. Now we're into a migration--they want the Sun to set. Sad, because like others I too was a huge fan of Solaris on Sun gear.
Can't bring myself to reccomend sparc/spolaris to anyone anymore.
with the explosion in support costs and lack of support, I recommend Sparc/Solaris to no one any more. while the hardware can be rock solid, and once you have the patches in place Solaris will run and run and run with little effort, essentially being forced to have hardware maint just to get software maint is just plain stupid. You are paying for the server twice in a 5 year or less period :( the first time at purchase, the second for the maint costs over 3-5 years. Unless you are one of their top customers have fun trying to get response out of them. $Work-1 went through an authorized reseller because trying to get a reponse out of sun once oracle started the process of trying to borg them became a nightmare.
and forget trying to get any patches anymore if you don't have a support contract.
Sun/x64 is just not competitive
I've been running Solaris/x86 in production for over 12 years, at first on Intel white box servers, then on Sun's excellent V20z and X4100 Opteron servers. Their x64 machines were competitively priced. In my new company I needed a quote for 40 servers, and they are now twice as expensive as HP, so we are going to switch. We'll still be running Solaris, but the OpenIndiana fork, since Oracle's support contract policy is unconscionable. And of course we ditched Oracle for PostgreSQL...
It's amazing, but Larry Ellison is even better at running Sun into the ground than My Little Pony was...
@ kebebbert. You really didn't argue with the facts. This is good, because they were correct. The personal abuse documents your skewed technique of manufacturing an analysis, attributing it to me and arguing with THAT.
- I guess that you wanted to argue that GHz doesn't bring linear performance improvements. I agree with that. But when a given technology increases its GHz, then things get somewhat faster. The z196 & P4 arguments didn't take major technology differences into account.
- I don't know the context of your Niagara comparison to Power 6. NOBODY thinks that Niagara is a good platform against Itanium, Power or x64 unless the workload is dramatically parallelized so that Niagara can deploy 10x the cores to compete.
- IBM Cell is a dramatically different animal from Power 7. It was developed on a parallel path
since Cell inception in 2001 (Power 3). It picked up die-shrinks & GHz upgrades, but not the 64-bit structure or I/O mechanisms that have been added since that time.
- I went back and checked my post. I clearly was stating the IBM dropped its price-structure in early 2010. Your response indicated that you agree, but felt that you disproved my comment. I don't get it. The Power 7 deployment in Feb 2010 significantly dropped IBM server costs. I think that we agree.
- Your Anandtech reference benchmarked a 15% difference and they ran the 16 month old Power 7 server when newer/faster/bigger servers OR newer/cheaper blades had already replaced it. I don't contest that Westmere 32nm is an impressive technology, it is. I do contest that enterprise deployments of large servers should abandon Power and move 100% to x64. IBM still has a cost structure that can favor Power when the environment is large enough and you look at the bottom-line (server + software + infrastructure + support). I feel that a sufficiently large shop still has a need for enterprise UNIX (and AIX).
I hope that helps...
As to Oracle...
All I can say is "Good!' I hope you lose a lot more money. Then perhaps it will occur to you to install a management team that actually knows that it is important to play 'nice' if you want to continue to make money. What they did with OpenOffice, for instance, is just a sad but well-known instance of sheer stupidity.
At this rate, Oracle will rank up there with Comcast as most hated company in the world. (or would that be Sony....?)
Blah blah blah
The article found one relatively small place that Oracle is not doing well -- Unix in eEurope, but of course neither are the other Unix vendors. On this same site it has been reported that Oracle is growing by over 30% in Unix world wide... Say again how terrible Oracle is doing? Just about every analyst out there is saying they were wrong and that Oracle was right.
This article stating that Sun is bringing down Oracle is total rubbish. Then to follow up with an article titled "Government spending freeze kills off spring server sales" just goes to show that the author obviously has ulterior motives.
OK! You don't like Oracle, nor Sun. Say it then! Even if you came out and said that everyone hates Oracle and Sun, and here's why... it would be better than this obviously skewed article. Two other articles stating how great Oracles HW business is doing, on El Reg of all places, and then this trash...
OK, I'm done.
- Updated Zucker punched: Google gobbles Facebook-wooed Titan Aerospace
- Elon Musk's LEAKY THRUSTER gas stalls Space Station supply run
- Windows 8.1, which you probably haven't upgraded to yet, ALREADY OBSOLETE
- Mounties always get their man: Heartbleed 'hacker', 19, CUFFED
- Android engineer: We DIDN'T copy Apple OR follow Samsung's orders