back to article Oracle to 'out-Sun' Sun on hardware and software

Oracle will out-invest Sun Microsystems on Sparc and Solaris and provide more open source contributions, Larry Ellison and Scott McNealy told OpenWorld Sunday night. Oracle's chief executive and Sun's chairman, joined by executive vice president of systems John Fowler, also made the pitch that Oracle customers should run more …

COMMENTS

This topic is closed for new posts.
  1. Destroy All Monsters Silver badge
    Welcome

    "The community would simply fork because the project is licensed under the GPL"

    Well, I don't know. There is this self-contradictory "dual licensing" scheme of MySQL whereby you are not supposed to use the GPL-ed version but the paid-for version in case of certain usage patterns (thus evidently contradicting what the license of the GPL-ed version says). Will lawyers have a field day?

    Where's the "welcome your lawyer" icon?

  2. TeeCee Gold badge

    A bell rang here, so a quick trawl gives:

    http://www.theregister.co.uk/2006/03/01/mysql_sap_certification/

    The last para of that one gains some additional amusement value now. It implies that SAP were keen on getting onto MySQL to prevent their customers shovelling cash into the pockets of their top competitor (Oracle) regardless of which way the contract went (Oracle on Oracle or SAP on, er, Oracle).

    The way things are going, it looks like Oracle will be left holding most of the cards here. A cynic might suspect that MySQL is a more important part of the SUN deal than Oracle are letting on.

  3. Matt Bryant Silver badge
    Happy

    Did Larry say that?

    Seeing as Sun have - in this decade - become synonymous with delayed and failed projects and products, outright lies ("We're not for sale") and empty rhetoric, and a record fall from market-leader to junk stock, maybe he should have said; "We're going to un-Sun Sun!"

    Interesting that he seems to be targeting IBM so much rather than Microsoft or hp. But then maybe that's because he knows the majority of his current Oracle installs are on Windows on ProLiant, and that no matter how much he puffs on about SPRAC-Slowaris, that just ain't gonna change.

  4. Anonymous Coward
    Linux

    @Matt Bryant

    OpenSolaris is not slow, are you thick?

    Do your research first before spewing your ignorance upon the comment section.

  5. Matt Bryant Silver badge
    FAIL

    RE: @Matt Bryant

    "OpenSolaris is not slow, are you thick?...." Seeing as I didn't mention OpenSolaris in my post, I would have to suggest that you are the challenged individual in this thread. Both Larry and I refered to the commercial SPARC-Slowaris, which is the offering where you pay to go slower than equivalent commercial UNIX offerings, not OpenSlowaris, where you get to go slower for "free" and usually on x64 kit not from Sun. Oh, and seeing as in my experience, both SuSE and Red Hat outperform OpenSlowaris on the same x64 kit for the same "free" price, you may want to reconsider your OpenSunshine bit about it not being slow. Well, you might consider it if you weren't so challenged.

  6. David Halko
    Happy

    @Matt Bryant -- HP Ripe For The Picking!

    Matt Bryant posts, "But then maybe that's because he knows the majority of his current Oracle installs are on Windows on ProLiant"

    That sounds like a ripe fruit, for the picking, considering that the fastest Intel compatible Oracle platforms are now Sun platforms.

    With Solaris, ZFS, and Flash - there is really nothing that Windows or any Linux platform can do to compete, until they get a real production file system. At this point, people might choose Oracle on Sun (best) or other category (everything else.)

    I am sure Adaptec will be releasing some very nice hardware with embedded flash, to compete in these other lucrative markets (i.e. Windows & Linux) to help those hold-over shops compete.

  7. Matt Bryant Silver badge
    Happy

    RE: @Matt Bryant -- HP Ripe For The Picking!

    Who shook Novatose's cage? Ah well, I suppose it's only to be expected, what with Larry starting a new generation of Sunshine myths, that the old Sunshiners will feel brave enough to resurface from where they've been licking their egos after the crushing news of the Sunset.

    ".....With Solaris, ZFS, and Flash - there is really nothing that Windows or any Linux platform can do to compete, until they get a real production file system....." <Yawn> Three points for you to consider, Dave. Firstly; you keep telling us ZFS and OpenSlowaris are "open", so if they really were that much in demand they'd be appearing a lot more on x64 kit from hp, IBM or Dell, when the reality is Slowaris on x64 is even the slowest selling flavour at Sun, where Linux on Galaxy sales outstrip Slowaris on Galaxy. And flash is a commodity technology and not unique to Sun, so again there is nothing to even suggest Sun has any advantage there. A clue for you Sunshiners - flash, either SSD or otherwise, is not a Sun invention.

    Secondly; there are other filesystems that do a better job than ZFS. BTRFS, Oracle's own (and not a rip-off of WAFL) is one for Linux, but both are dwarfed by the number of Windows Server installs. You may have noted that Windows Server marketshare hasn't been affected in the slightest by over a year of fevered hype and frothing from you Sunshiners? That's because the vast majority of customers have zero interest in ZFS or Slowaris.

    Thirdly; hp ProLiant is the market-leader for x64 for a reason - because hp bring the right products, innovations and services to market at the right price. Sun's Galaxy is a "me too" market-trailer for another reason - Sun have never had the expertise, services, investment or appeal to compete. Just look at the blades market and then try and tell me Sun's x64 is anything other than a third tier effort. Even if Larry poured money into Sun's hardware bizz, he'd still have to spend a phenominal amount to catch up with Dell, let alone hp. And Larry is looking to make money out of that Sun hardware bizz (or so he says), not pour money into it. I feel pretty comfortable in predicting that the Soracle x64 products will still be "me too" market-trailers for years to come.

    /SP&L

  8. Anonymous Coward
    Thumb Up

    Surprisingly Good Move

    Knowing from experience that most high-end sales are actually driven from business units that are IT customers, this is actually a good place for Larry to be making his pitch.

  9. Jesper Frimann
    Big Brother

    Re:@Matt Bryant -- HP Ripe For The Picking!

    Ehh... ?

    So you honestly think that putting a 'We know best' filesystem between the database and the disk is going to get you a very good performance ?

    I think you have been reading the wrong blogs at SUN, like:

    http://blogs.sun.com/roch/entry/zfs_and_directio

    Man the guy must have hit his head, or haven't worked much with databases. Luckily there are more sensible guys like this one:

    http://blogs.sun.com/realneel/

    Who actually know what he is talking about and makes sure that ZFS does not stand in the way of the database. Which is what you should do.

    ZFS is way to hyped IMHO.

    // Jesper

  10. Stu 6
    Coat

    A little off-topic here...

    Is it just me or does anyone else notice a resemblance between the picture of Larry Elison and Tony Stark (Robert Downy Jr.) from Ion Man? Perhaps... Larry really is "Big Iron" Man!!

    Mine is that really jazzy Hot-Rod Red and Gold number over there...

  11. Jon999

    @Stu 6

    Well as the rumour goes, Tony Stark in Iron man(the film that is, not the comic) was modelled after Larry Ellison.

    Not sure if that is true, but when you go through the film you could imagine that they did model it on him

  12. Anonymous Coward
    FAIL

    Matt Bryant can't spell, again.

    poor mattie - can't spell SPARC or Solaris again

    poor unemployed mattie hates that OpenSolaris is consistently faster than multiple versions of Linux on the same hardware

    http://flexense.com/resources/nas_performance_comparison.html

  13. zvonr
    Thumb Down

    @Matt

    BTRFS does a better job than ZFS ?

    Can you name one thing it does better? (besides crashing and loosing data)

    BTRFS is not production ready and Oracle's future interest in developing it is highly questionable.

    If Redhat wants to avoid being replaced by Solaris, it needs to throw in some money at BTRFS development. Oh wait but then Unbreakable Linux gets it too, and Suse Linux too.

    So what is left is to create the "Aliance of the willing" and we will get something similar to systemtap.

  14. Matt Bryant Silver badge
    Happy

    No, Slowaris ain't slow.....

    Seeing as the Sunshienrs are back to regurgitating their old, tired excuses, evasions and insults, let's take a quick trip down memory lane to the last time Novatose tried cherry-picking benchmarks (http://www.theregister.co.uk/2009/02/27/hp_sun_oem_comment/comments/), and how Slowairs on Sun's x64 compared to SuSE on ProLiant:

    ".....But for real fun let's look at some more recent Opteron-based hp bench sessions. How about last December's 2008064, where the hp ProLiant DL785 G5 8-socket posted 35400 SAPS, compared to the October 2008 bench session 2008061, where the similar Sun Fire X4600M2 8-socket only managed 29670 SAPS. Yes, I know the Sunshiners will start whining about how the hp server had the 2.7GHz 8384 4-core Opterons, whereas Sun's server had to make do with the 2.5GHz 8360 4-core Opterons, but it's strange that 8% jump in CPU clock gave a 20% jump in performance. Oh, hold on a sec, I see the differentiator - the Sun box was struggling with Slowaris 10, whereas the DL was humming along with SuSE Linux...."

    /SP&L

  15. David Halko
    Go

    RE: RE: @Matt Bryant -- HP Ripe For The Picking! #

    Matt Bryant posts, "you keep telling us ZFS and OpenSlowaris are 'open', so if they really were that much in demand they'd be appearing a lot more on x64 kit from hp, IBM or Dell..."

    Your spelling is atrocious. The product you couldn't spell is "OpenSolaris".

    Your word search capability is also deficient. No one in the threads before your post ever made any argument indicating that the ZFS or OpenSolaris were "open". But since your post illustrates confusion over this, here is a document which details how you can also contribute!

    http://www.opensolaris.com/use/SourceJuicer.pdf

    Your connection with something someone didn't say to non-cited competitive vendor adoption rates is nonsensical. Your argument (linking "open"-ness and hardware competitor "demand") deteriorates by the fact that Microsoft Windows Server holds significant market share.

    Matt Bryant posts, "the reality is Slowaris on x64 is even the slowest selling flavour at Sun, where Linux on Galaxy sales outstrip Slowaris on Galaxy."

    Your spell check failed a couple more times, again. The product name you can't spell is "Solaris"

    The HPC clusters really drive the Linux sales at Sun. Most substantial HPC clusters use Lustre.

    http://www.hpcwire.com/offthewire/Suns_Lustre_File_System_Powers_Top_Supercomputers.html

    Once Lustre is ported to Solaris, Solaris will be competitive in this area, since HPC clusters don't normally leverage commercial clustering software like QFS. More Lustre features are arriving in ZFS with every Solaris 10 release, so it is just a matter of time.

    Matt Bryant posts, "flash is a commodity technology and not unique to Sun, so again there is nothing to even suggest Sun has any advantage there"

    All flash is not created equal. Some flash is faster and less reliable, some flash is slower and more reliable.

    What makes Sun's advantage so aggressive in the market is the intelligent use of the each kind of flash, in ZFS at the kernel layer, to provide transparent application acceleration. Failure of flash cache can be tolerated (and locked out) while failure of flash file system intent logs is less tolerable, should be high quality, and should be mirrored.

    Matt Bryant posts, "there are other filesystems that do a better job than ZFS. BTRFS... is one for Linux"

    http://btrfs.wiki.kernel.org/index.php/Main_Page

    "Btrfs is under heavy development, and is not suitable for any uses other than benchmarking and review. The Btrfs disk format is not yet finalized, but it will only be changed if a critical bug is found and no workarounds are possible."

    Unreliability is a key focus of Matt's Linux postings. Looks great on benchmarks and in slide-ware. (Even the default configuration of most Linux NFS configurations is to ignore NFS client write sync to disk on the server side.) But, that's OK - Sun sells & supports Linux for those crowds, too!

    Matt Bryant posts, "hp ProLiant is the market-leader for x64 for a reason - because hp bring the right products, innovations and services to market at the right price. Sun's Galaxy is a 'me too' market-trailer"

    HP targets a different market than Sun with their Intel compatible line. I guess HP could make ProLiant more competitive with OpenSolaris, ZFS, and using some nice HP USB thumb drives (for ZFS cache) tied next to their USB printers. There is nothing wrong with that, it is just a different focus area. ;-)

    Sun has targeted the Galaxy line at HPC systems, expanded the focus to network appliances for very large scale media storage installations (i.e. video, audio, news, etc.) and Oracle, Postgres, and MySQL appliances & implementation support. Sun has been consistently scoring superior or competitive in these areas in relation to the competition.

    With the flash storage system integrated into ZFS File System, ZFS integrated into the OpenSolaris and Solaris Kernel, integrating CIFS directly into the OpenSolaris Kernel, creating awesome InfiniBand switched infrastructure, embedding 10GigE into the CPU chips on some of their offerings, and open sourcing Solaris hardly makes the product suites a "me-too" offering from Sun.

  16. David Halko
    Go

    RE: RE: @Matt Bryant -- HP Ripe For The Picking!

    Jesper Frimann posts, "So you honestly think that putting a 'We know best' filesystem between the database and the disk is going to get you a very good performance?"

    The use of flash for the ZFS Intent Log and L2ARC cache seems to provide very good performance.

    Considering it is end of 2009, going back to a benchmark for a 2006 version of ZFS (when ZFS originated in 2004) is a relatively poor way to understand performance and dynamics of a file system. The changes and features released with ZFS are absolutely astounding!

    http://netmgt.blogspot.com/2009/10/zfs-next-word.html

    Performance is not the only need behind a database. I have seen your posting before, I know you know this. Life cycle investment is also a key consideration. There are administrative efficiencies, migrations, replications, backups, etc. that ZFS makes easier.

    Having a system admin swapping smaller disks for larger disks to automatically expand the size of a zpool is far easier & less resource intensive than LUN migration in a SAN system by application, storage, and DBA teams. The database can also efficiently manage the raw disk slices directly of disks added to a system, but do I want a DBA working with a system admin by adding raw slices in a datafile to a tablespace, to get a little more efficiency?

    I would rather pay a DBA to optimize SQL in a database rather than to add some disks for storage. Good DBA's are hard to find, and using their time wisely is a better use of resources.

    Every application or file system requires a bit of tuning, to get the most out of it, depending on the workload type. The blogs you point out demonstrate that with ZFS, you can have it all.

  17. Anonymous Coward
    FAIL

    RE: @Matt Bryant -- HP Ripe For The Picking!

    Matt Bryant Posted Monday 12th October 2009 14:52 GMT

    --- Sun's Galaxy is a "me too" market-trailer for another reason - Sun have never had the expertise, services, investment or appeal to compete.

    Is that why Oracle kicked HP out on their backsides??? LOL!!!

    http://www.oracle.com/us/corporate/press/036544

    "“With this benchmark result, there’s no denying that Oracle Database 11g running on Sun SPARC Enterprise T5440 servers outperforms IBM and DB2,” said Juan Loaiza, senior vice president, Systems Technology, Oracle."

    “No other vendor today is shipping fully-integrated flash-based hardware and software that leverages a world-class operating system - Solaris - to deliver these breakthrough world record performance results,” said John Fowler, executive vice-president, Systems Group, Sun Microsystems.

    If Oracle asked HP to buy Sun hardware, HP said no, HP lost the Exadata deal, and Sun hardware is now the highest performing database system - it sounds like HP did not have the expertise and refused to invest!!!

    ROTFL!

  18. Captain Thyratron

    Youth these days.

    There used to be a "Don't Feed The Trolls" sign hammered into the ground next to the pond, but some hooligan seems to have absconded with it. Better check the local pawn shop.

  19. Anonymous Coward
    FAIL

    mattie 'tard and last time reading his stupid reference...

    Matt Bryant ~~~ Seeing as the Sunshienrs are back to regurgitating their old, tired excuses, evasions and insults, let's take a quick trip down memory lane to the last time Novatose tried cherry-picking benchmarks (http://www.theregister.co.uk/2009/02/27/hp_sun_oem_comment/comments/), and how Slowairs on Sun's x64 compared to SuSE on ProLiant:

    Matt accused people of posting items they didn't post, when following the reference. HA HA HA!

    This Matt guy can't even spell sunshiners, Solaris, OpenSolaris, SPARC. LOL!

    Matt didn't even know how to use the phrase "hoist by your own petard"! ROTFL!

    Matt talked claimed a quad-core chip was a dual core chip! LMAO!

    Matt even cites an old thread where he admits to being unemployed with no skills or experience. NOOB!

    The article was about HP Itanium having to deal with Intel and being an OEM of Solaris 10, yet Mattie just didn't have the capacity to discuss HP & Itanium, and instead turned it into a hating-other-systems session. WHAT A TRIP!

    Matt falls back to hypocrisy, hurls insults, and accuses others of it, in the reference he provided, as well as in the current thread. (kind of sad, though - so let's see what he has to say about a benchmark dating back a year or so, out of pity for the unemployed illiterate NOOB)

    Matt Bryant ~~~ (various drivel comparing SAPS Unicode benchmarks on Sun's Solaris 10 to non-unicode benchmarks from HP running Novel's SUSE)

    http://www.sap.com/solutions/benchmark/sd.epx

    "Unicode and the use of the new general ledger are now common practice for SAP customers across all industries, and the SAP standard application benchmarks need to reflect this change. The updates are transparent; that is, the steps of the benchmark scenario remain unchanged. Please be aware that these changes make the SD benchmark more resource-intensive, which has a direct impact on the benchmark results."

    Now that Matt showed himself to be a complete fool, for the SECOND time, by reminding the world AGAIN that he did not understand the benchmarks that he posted - I am about done bothering to consider ANYTHING that he says.

    It was an interesting "trip down memory lane". It is a trip I won't bother to take again. Matt is just a waste of time.

    FAIL because Matt Bryant failed, a second time - once was not enough.

  20. Anonymous Coward
    Thumb Up

    re: mattie 'tard and last time reading his stupid reference...

    That's hilarious. I went back and looked at the post that Matt referenced. He did indeed admit to being unemployed and inexperienced. Amazing. We really do need to stop feeding this troll.

This topic is closed for new posts.

Other stories you might like