Forgot your password?
typodupeerror
Oracle Unix Upgrades Hardware

Oracle's Sparc T5 Chip Evidently Pushed Back to 2013 98

Posted by timothy
from the when-visions-collide dept.
Mark Hachman writes in Slash Datacenter that the Sparc T5 chip Oracle announced earlier this year apparently won't be ready until sometime in 2013. John Fowler, executive vice president, Systems, Oracle, presented at Oracle Open World a chart outlining highlights of Oracle's plans for the future. "But Fowler also skipped over some bad news: an apparent delay for the Sparc T5. A year ago, Oracle’s Sun division announced the Sparc T4—and according to Fowler, Oracle chief Larry Ellison set a very high bar for the next iteration: double the performance while maintaining app compatibility on an annual basis. Apparently, that didn’t quite happen with the T5; Oracle had the opportunity to announce a T5-based server, and didn’t. That’s a bit of bad news for the Sun design team, which already had to watch Intel’s Xeon chief, Diane Bryant, give the preceding keynote. ... As detailed at this year’s Hot Chips conference, the T5 combines 16 CPU cores running at 3.6 GHz on a 28-nm manufacturing process. Continuing the trend of hardware acceleration of specific functions, Sun executives claimed the chip would lead in on-chip encryption acceleration, with support for asymmetric (public key) encryption, symmetric encryption, hashing up to SHA-512, plus a hardware random number generator."
This discussion has been archived. No new comments can be posted.

Oracle's Sparc T5 Chip Evidently Pushed Back to 2013

Comments Filter:
  • Oracle? SPARC? (Score:4, Interesting)

    by dyingtolive (1393037) <brad.arnett@notforhire3.14.org minus pi> on Thursday October 04, 2012 @05:42PM (#41553403)
    Someone is still buying that shit?
    • Re:Oracle? SPARC? (Score:5, Interesting)

      by fm6 (162816) on Thursday October 04, 2012 @06:13PM (#41553597) Homepage Journal

      Before Fowler became an Oracle employee, he was in charge of the hardware division at Sun. And before that, he was in charage of x64 systems. I was working there at the time, and the word from on high was that putting the x64 guy in charge was a signal about our future direction.

      Which of course, didn't happen. Sun's sales channels continued to view x64 systems as a way of migrating people to SPARC vis Solaris-on-x64. Which all our customers, who were already heavily invested in Windows and Linux, had no interest in. My big hope for the Oracle takeover was that Oracle's sales org (aside from being bigger than all of Sun) would be smarter than that and push x64 systems.

      But Oracle has dratically reduced the models of x64 systems they sell. Officially, that's about a leaner product line and ending the special relationship with AMD. But I'm beginning to expect that the SPARC koolaid is as popular in Oracle as it was in Sun.

      • Sad that they aren't going the opposite direction. As a software vendor, our company is working (granted, in glacial terms) towards phasing out support on SPARC platforms.
        • Re:Oracle? SPARC? (Score:4, Insightful)

          by fm6 (162816) on Thursday October 04, 2012 @06:50PM (#41553895) Homepage Journal

          Glacial indeed, if they haven't already done it. Like the other 99% of the industry.

          One has to be really dense not to see this trend. ALPHA is gone. MIPS is only used in embedded devices. Itanium and POWER are strictly legacy products. And yet people still believe that SPARC can survive in the server space.

          I'd be sad too if I still worked at Sun. But not only does the failure of this product line no longer affect me, even abandoning SPARC completely would not save it. Computers are Dead [slashdot.org].

          • by Anonymous Coward

            I'm not sure about that, but one big issue with power, itanium and sparc is: NONE OF THEM HAVE LOW END SYSTEMS TO SPUR NEW DEMAND.

            I mean fuck, seriously? If each of those archs had a current-generation workstation, nothing fancy, a couple cores, probably with basically no cache and such compared to the high end models (maybe even just reject bin parts from the high end systems), a decent selection of PCIe and maybe a few PCI slots, and run inside a standard 120V 15A circuit, I'd look into buying one for up

          • by Anonymous Coward

            I'd like to hear about how POWER is a strictly legacy product, especially since POWER 7+ was released today and future generations of the processor are in the design and production pipeline.

            • And the new PureFlex systems are actually pretty decent hardware if you are in a market for a scalable "blade" system that is not really a blade.

              The power management abilities of the hardware are pretty decent, allowing computing power to spin up and down depending on demand.

              Yeah, a lot like a standard PC or laptop right? but on a larger 100+ socket scale. Power and cooling are significantly less than stand alone systems or blade chassis.

          • Re:Oracle? SPARC? (Score:5, Interesting)

            by Phat_Tony (661117) on Thursday October 04, 2012 @09:43PM (#41554903)
            Five years ago your comment would have made a lot of sense to me, but now you're talking about how everyone's gone X86 during the first massive movement away from X86 the industry's seen... smartphones and tablets are all computers that run on ARM processors, they're cleaning X86's clock in the only rapidly expanding market. And ARM's next core design is aimed at servers.

            For the first time, Windows compatibility is mattering less and less as many users only use the web and web apps on their computers - opening the door to competing processors for the first time since the late 80's. At the same time, PC's continue to represent a smaller and smaller share of new CPU's, which are migrating to data centers, smartphones, and pads, which are even less dependent on X86 compatibility.

            For the first time, the computational penalty of X86 instruction set translation for RISC cores may not outweigh the compatibility benefit for a significant portion of users. Increasingly, customers don't care about compatibility with existing X86 codebases. Like ARM, anyone with a new processor with compelling performance per watt might actually be able to sell the thing, without everyone assuming it's worthless if it won't run Windows.

            Also, I wouldn't quite characterize POWER as a strictly legacy product, since IBM introduced the latest iteration, the power 7+, in August 2012, and is currently selling 15 different systems using Power7 processors. Not to mention the Xbox 360, Playstation 3, Wii, and not-even-out-yet Wii U that are all POWER based systems.
            • by fm6 (162816)

              Yeah, my obsession with the x64 versus everything else war is becomming less and less relevant. Mobile devices are indeed about ARM (though Intel would like to change that). And I agree with you that web applications are chiping away at Windows dominance.

              But ARM servers? People have been trying to sell those as long as ARM has been around. ARM advocates are insisting that the latest improvements will give them the edge they need, but the factors that keep the data centers full of x64 systems have not change

              • AIX runs on Power7. There also is RHEL for Power7:

                http://www.redhat.com/products/enterprise-linux/for-ibm-power/ [redhat.com]

                However, AIX does not run on Mainframe hardware, although you can run Linux on the Mainframe.

                So you can run Linux on anything IBM sells, but that is the only OS since OS2 to do it.

                • by fm6 (162816)

                  I guess you can probably run Linux on anything designed in the last 20 years. But how many people buy Power 7 systems to run Linux? Very few, I suspect.

                  Dumb of me to classify AIX as a mainframe OS. But it's still a legacy OS.

                  • I work for IBM, in the Midrange space, (and if it is not Mainframe or AS/400, it is Midrange, unless it is a embedded system/appliance) so I see tons of it coming through the pipeline, especially when the customer wants an enterprise sized install, like a Data Warehouse.

                    Fast as x86 CPU's are, if you need I/O speed, it is still RISC systems, including SUN and HP.

                    AIX is still IBM's premier UNIX based OS for the enterprise. RHEL is Linux solution. from Enterprise to desktop.

                    Having SA'd using HP, *BSD, AIX, SC

                    • by fm6 (162816)

                      AIX is still IBM's premier UNIX based OS for the enterprise.

                      I'm sure it's a great OS. But is anybody buying it? There's certainly no place for it in the cloud-oriented data center.

                      Anybody who's worked in the computer industry for any length of time knows that "great technology" is not the same as "stuff people want to buy". Computer history is littered with the corpses of products that were technologically wonderful but which couldn't find a market.

                      AIX was launched in 1986. This was a very fortuitous time; on the one hand, it was a modern Unix-style (being Unix!) OS

                    • by Anonymous Coward

                      IBM sells about 5 billion dollars of AIX-running hardware per year.

                      Next question.

                    • Again with the ignorance.

                      We sell more AOD to customers than other cloud offerings. Application on Demand gives you Power+ running AIX or Linux on IBM's cloud infrastructure, accessed remotely.

                      The new PureFlex architecture allows you to mix AIX and Linux on Power+ using the same cloud provisioning toolset across both platforms.

                    • by fm6 (162816)

                      So IBM platforms are big on the IBM cloud? Why am I supposed to be impressed by that? The measure of a platform's importance to cloud computing is its role in all vendors.

                      And that "ignorant" is rude and uncalled for.

                • What exactly did OS/2 run on other than PCs? POWERstations? Not that I recall. AS/400s? Nope! What else could it run on?

                  For the POWER line-up, there is Linux, and then, there is BSD - I believe that all the big 3 are supported there. So IBM has a choice of at least 4 OSs to put on POWER, aside from AIX. It's a shame that OS/2-PPC never happened, but maybe osFree could happen, if the L4 microkernel is ported to POWER, and then POWER could run a Presentation Manager based OS.

                  • Good gods.

                    OS2 ran on PC, PowerPC/RS6000 and mainframe.

                    AS/400 was migrated to RS64 architecture in 96, so you could run AS400 on it. AS400 became an OS, not a hardware line. Under the covers, it was the RS6000 line.

                    • by unixisc (2429386)
                      When did OS/2 run on the PPC or RS/6000? A different OS called Workplace OS - which was Presentation Manager on top of Mach 3.0 was developed, but IBM pulled the plug on it before it was ready. Yeah, AS/400 was migrated to the RS/6000, but unfortunately, never OS/2. Otherwise OS/2 on PPC, w/ full IBM backing, might have been a great hit, and might even have attracted the Mac clonemakers like Power Computing, Motorola CPG and Umax to switch to them once Jobs cancelled the MacOS licensing.
              • But ARM servers? People have been trying to sell those as long as ARM has been around.

                No they haven't. ARM2 was definitely a mobile / desktop chip. Up until AMR6 they were mainly aiming at this same market, and then focussed almost exclusively on embedded. Cortex A8 was the first that anyone seriously thought about putting in a server, but ARM wasn't pushing it in that direction. Cortex A15 is the first chip that they've designed aimed at servers and we're only just seeing shipping silicon for it now (and most of that is aimed at tablets). ARM has definitely identified the low-power ser

                • by fm6 (162816)

                  ARM2 was definitely a mobile / desktop chip.

                  It was used as a desktop chip. (Don't recall a lot of mobile devices in 1987.) What prevented it from going into a rack-mount server? Lack of customers, I should think.

                  A lot of traditional file server roles in small businesses have been replaced by ARM-based NAS appliances over the past few years,

                  Yep.. I own a QNAP.TS-110. Sweet little box.

                  • It was used as a desktop chip. (Don't recall a lot of mobile devices in 1987.)

                    You might remember the Apple Newton. The reason ARM was spun out from Acorn was that Apple wanted to use their CPU, but didn't want to buy it from a direct competitor. Having a spin-out that provided CPUs for both of them was fine.

                    What prevented it from going into a rack-mount server?

                    No one tried it? It simply wasn't a market ARM aimed at. The two operating systems that ran on it were NewtonOS and RiscOS, neither of which was aimed at servers. They could have ported something else, but Linux didn't exist for 4 years and wasn't really credible for a decade

              • by wireloose (759042)

                . But those of us who are willing to wait a year or two for the latest GTA to be ported to the PC just don't care.

                Good point, but that's more of a consumer view than a business view. Businesses are usually working on the here and now, or the near future, and cost effectiveness. Your view leans toward less timeliness to gain the cost savings. I have both views, one for my personal equipment, the same as yours, and one for my company, the timely need.

                You also make some good discussion about x64 compatibility, but consider services and apps vs. hardware. If you're connecting to my web site or my cloud services, you do

                • by fm6 (162816)

                  If you're connecting to my web site or my cloud services, you don't know what hardware I'm running on, but you DO care that it's fast enough to meet your needs. So why would you care whether my hardware is x64 compatible as long as your x64 systems talk to it just fine?

                  Of course I don't care. But I'm not the person who's building the infrastructure that makes this web application work. And that person wants commodity systems: lower upfront cost, lower TCO.

                  IBM Power systems also run Linux.

                  As I already pointed out, everything runs Linux. But how many people are buying POWER to run Linux?

                  Well, at least IBM is trying to push Linux on POWER itself. At Sun, we left Linux on SPARC to Canonical. But I don't see either taking off any time soon.

                  • by wireloose (759042)

                    If you're connecting to my web site or my cloud services, you don't know what hardware I'm running on, but you DO care that it's fast enough to meet your needs. So why would you care whether my hardware is x64 compatible as long as your x64 systems talk to it just fine?

                    Of course I don't care. But I'm not the person who's building the infrastructure that makes this web application work. And that person wants commodity systems: lower upfront cost, lower TCO.

                    Commodity systems do not always mean lower TCO. TCO is based on a lot more than just hardware, which is normally a small fraction of a system's costs.

                    IBM Power systems also run Linux.

                    As I already pointed out, everything runs Linux. But how many people are buying POWER to run Linux?

                    Well, at least IBM is trying to push Linux on POWER itself. At Sun, we left Linux on SPARC to Canonical. But I don't see either taking off any time soon.

                    Very well, but you also said of IBM, "When they start selling POWER systems that run Linux, then we can talk." I was merely responding to that, fyi. I'm certainly not putting words in your mouth.

                    • by fm6 (162816)

                      TCO is based on a lot more than just hardware, which is normally a small fraction of a system's costs.

                      Oh, is that why they call it "total cost of ownership"? Gee, that never occurred to me!

                      The post-purchase costs of non-standard systems are pretty substantial. I found that out first hand when I tried to be a good Sun employee and run my internal wiki on Solaris instead of Linux. Kept running into TWiki plugins that didn't work Solaris because they depended on Perl libraries that had only been tested on Linux.

            • Re:Oracle? SPARC? (Score:4, Interesting)

              by serviscope_minor (664417) on Friday October 05, 2012 @05:42AM (#41556725) Journal

              For the first time, the computational penalty of X86 instruction set translation for RISC cores may not outweigh the compatibility benefit for a significant portion of users.

              Yes and no.

              Yes, in that it's always mattered. Intel can never be power competitive in the low end due to the expense of the x86 instruction decoder. Then again, neither can ARM which is why static 14/8 bitters like PIC dominate the truly low end. In the mid range ARM and others (but mostly ARM) will not be displaced by Intel for exactly that reason.

              ARM dominate all the way to the beginning of the high end. However, once one hits the high end, and single thread performance goes up, again the decoder becomes a smaller and smaller fraction of the energy usage. The OoO unint and execution units dominate, as the OoO unit has to expend a lot of energy to keep the energy hungry execution units fed while they're ungated.

              Users are now beginning to care about the high end on their phones, just about.

              For fun, compare the FLPOS/Watt of an i7 Ivy Bridge to any other general purpose CPU. The Ivy Bridge one does surprisingly well, in fct I think it's pretty much a winner. Certianly compared to ARM. The reason is that as the performance goes up, the instruction decoder begines to pale in to insignificance.

              It's the same old argument as always.

              Intel will never hit the mid to low end, but the penalty almost disappears on the high end, and Intels better process and expertise in branch prediction dominates.

              For now, phones are bumping into the bottom of the top end. In 5 to 10 years they will be firmly in it, and the landscape will be very different. In 10 to 15 years, cheaper smartphones (e.g. spiritual successors to something like the Galaxy Ace) will be comfortably inot the high end.

              Arm will continue to dominate the upper low end to the top of the middle because of the decoder. But phones things will be moving well into the top end.

              I'm defining low/middle/high by absoute performance and relative power tradeoffs between parts of CPUs.

            • they're cleaning X86's clock in the only rapidly expanding market ... And ARM's next core design is aimed at servers.

              it's important to qualify that. ARM uses a lot less power, but it's not more power efficient. it uses less power because it's so much less powerful.

              as much as mobile mobile has been ARM's domain, servers are intel's domain. making a claim that they are "aiming at servers" is one thing, toppling the giant is quite another.

              as much as i'd like to see intel have some healthy competition, in all likelihood they'll end up steamrolling ARM with their superior manufacturing technology and their deep pockets, in the

          • by unixisc (2429386)

            Alpha & PA-RISC are indeed gone. Itanium too is almost there, although it could still have a life in supercomputers. SPARC's real strength was in the workstation space, as well as the database server space of Oracles. Indeed, there was a time when SPARCstations were ubiquitious for CAD engineering work, such as Cadence, Verilog and the HDLs.

            POWER pretty much owns the games console market, having conquered it from MIPS. MIPS now has the router and tablet markets, the latter where it's a better alte

            • by fm6 (162816)

              it would be nice if POWER returned to Apple as well - both iPads and Airbooks.

              Do you see that happening? Apple must have had a lot of motivation to make the painful transition from POWER to commodity. The transition back would be just as painful, and I don't see their motivation.

              x64 owns the desktop and the data center. ARM owns the mobile space. Anybody who thinks that's going to change any time soon as a bad case of the if-onlys.

    • Re: (Score:2, Interesting)

      by Anonymous Coward

      Wow. Have you actually used Oracle for a data need that actually requires it? How about a T4 SuperCluster? No? Really? That's probably why you're convinced that your comments like "Someone is still buying that shit" and the like are just hilarious. They're not. In fact they show your immaturity and lack of understanding of just how large and complex some data-sets can get. We've got racks of Exadatas being fed by racks of Superclusters and backing up to racks of ZFS backup appliances. We've also go

      • No, I haven't used Oracle for a data need that actually requires it. I'm not aware of what a data need that would actually require it would look like. I do work on high speed market data delivery systems. Top end stuff for a internationally relevant company (can't say who, but it's not Bloomberg). Mining databases doesn't really factor into the equation. In fact, all of our stuff is moving away from SPARC simply because we need inexpensive powerful processors, not thousands of threads all running at 1.
    • government agencies, all the better to track you. And oracle financials is popular in the business world. next iteration we are probably going to sparc not because we want to but because of crappy oracle contracts. yes it is a shit solution and a shit product, but who's running to great plains now a MS product...????? trapped, that is what we are. :(
    • Processors are very much a Yoda situation of "Do or do not, there is no try." For high end servers, there is a market for non-x86 stuff. However to be in it, you need to be up on the curve, you need to invest real resources in development. On the other hand you can just get our of it and buy product form someone else, probably Intel but IBM or Hitachi are options. It is expensive, deciding not to compete is 100% valid.

      However half-assing it is going to lead to nothing but wasted money. You can't decide you

  • T5 combines 16 CPU cores running at 3.6 GHz on a 28-nm manufacturing process

    WOOOOO! Impressive!


    ...
    eh, not really...
    What can Oracle offer with the "T5" that isn't offered better, faster, and/or cheaper on Xeon or Power?

    Oh yeah, I forgot, Oracle can offer "register windows"... Wheeee!

    • by Anonymous Coward

      T5 combines 16 CPU cores running at 3.6 GHz on a 28-nm manufacturing process

      WOOOOO! Impressive! ...
      eh, not really...

      What can Oracle offer with the "T5" that isn't offered better, faster, and/or cheaper on Xeon or Power?

      Oh yeah, I forgot, Oracle can offer "register windows"... Wheeee!

      How many registers in the x86 set? 6 32-bit. You can get 14 if you're running 64 bit, and 8 double-precision (actually 80-bit) FP.

      How many registers in the SPARC v9 set? 31 64-bit general purpose, 32 double-precision/16 quad-precision FP.

      And one die of this chip will offer 8 CPUs, each capable of running 8 threads simultaneously for a total of 64 threads/die.

      What's Intel up to? 8 CPU/die with 2 threads/CPU for 16 threads/die?

      • Re: (Score:3, Insightful)

        "If you were plowing a field, which would you rather use: Two strong oxen or 1024 chickens?" - Seymour Cray.

        The devil is in the details. SPARC has lots of registers, very true. But it needs more user-accessible registers, because its address modes are simpler, and you need to do more address computations in registers. Register windows were like a fully associative cache for a few levels of your call stack... but then you have to save more stuff when you do a context switch, and I suspect they were part o

  • What are these chips used for?

    Oracle still makes sparc servers but no sparc workstations as I know of. So, do these chips get somewhere else? And the following question is why one would want a sparc server in the first place?

    • by timeOday (582209)
      It's a good question but if any market can justify complete vertical integration including customized processors for a specific application, it's Huge Databases. (But I did say if).
      • As someone else said [slashdot.org], "ALPHA is gone. MIPS is only used in embedded devices. Itanium and POWER are strictly legacy products. And yet people still believe that SPARC can survive in the server space."

        The cpu markets is headed to support only two kinds of microprocessor: ARM where there is restrictions on power consumption and x86-64 everywhere else. Is there really a viable market for specialised CPUs?

        • by cpghost (719344)

          Is there really a viable market for specialised CPUs?

          Last time I've used them, SPARCs were general purpose CPUs... and pretty good ones at that. But yes, there's a market for that. For instance, Fujitsu SPARC64 CPUs are currently being used, among others, in the HPC world [theregister.co.uk] for massively parallel simulations.

    • by Anonymous Coward

      Think about it. Who bought Sun?

      What's Oracle's biggest product?

      What's the most cost efficient strategy for scaling Oracle DB?

      Compare the cost of a 32-core (supporting 256 simultaneous threads) Sun machine to the competition?

      Not so steep, now, is it?

  • The article is a link to another slashdot article. Seriously, what the fuck? If you cared, shouldn't you just comment in the original article rather than here? Or post the full text here instead of a summary of a link to another page on slashdot?

    At least you're not posting comments as stories. Yet.

    Maybe the Dice overloads will get rid of this dumb shit?

  • by ubergeek65536 (862868) on Thursday October 04, 2012 @07:13PM (#41554031)

    "hashing up to SHA-512" How many MH/s per Watt? Does Oracle take bitcoins?

  • by Anonymous Coward

    I recently bought a Sun Fire T1000 so I could play with Solaris. I found out the hard way that the default logical domain manager in Solaris 11 doesn't support T1 processors. I went to downgrade to the previous ldom software just to find out that software is behind a paywall.

    I could use Solaris 10 instead, but I'm screwed out of using their latest and greatest software because of their paywall. I also can't update my firmware due to the same reason.

    Oracle, do you really expect new customers to buy your p

    • Yeah, do not buy old Sun hardware thinking that you can get any useful support from third parties, or pick up a cheap support contract suitable for a sysadmin's home box or a dev workstation... or even download firmware for a device that is not covered by your current support contract. That sort of thing went away by or shortly after the time that Oracle bought Sun.

      Oracle doesn't really care about ISV support for SPARC, and they probably like it if their big Oracle/SPARC sales included a hefty dose of high

      • by eric2hill (33085)

        The firmware thing was what caused me to start recommending other server manufacturers. The Sun hardware was actually really nice, well designed, and very stable. The ILOM was great since it was so tightly integrated with the hardware and yet completely out of band, and was included with the server at no real additional cost.

        Then Oracle bought Sun and turned off firmware support unless you had an active support contract. That was a big *fuck you* to everyone who bought a bunch of Sun hardware and only ke

    • by laffer1 (701823)

      It's not a waste. You should be able to run a Linux or BSD on it. Your choices are limited, but you don't have to run solaris. You may also be able to get an open source fork of opensolaris to run on it.

  • Man, after all of Sun's customers got screwed by Oracle buying them I cant believe people are still investing in Sparc hardware. Sun''s marketshare has gone from 16% in q1-2008 to 4.7% in q1-2012 as HP, IBM and Dell's marketshare have increased. With declines like that; its only a matter of time before Oracle turns away the rest of their server customers by killing off Sparc an Solaris.

    • by unixisc (2429386)
      Well, Oracle responded by killing its apps on Itanic, and could easily do the same to POWER if it wanted. In which case, SPARC would be their only non-x86 platform
      • by yuhong (1378501)

        Or tried to. This attempt led to a lawsuit from HP.

        • by unixisc (2429386)
          Well, they effectively won. Anybody who wants to build their high end solutions around Oracle software now knows not to go w/ Itanium, since Oracle doesn't necessarily have to provide the latest or greatest - they can even provide mediocre support. In the meantime, anybody who does want high end Oracle based solutions is much safer going w/ UltraSPARC or Sparc T5 based solutions than even POWER, MIPS or even Itanium.
  • The Itanium division at intel was so overjoyed they ordered a medium pizza, with wings to celebrate...

"You know, we've won awards for this crap." -- David Letterman

Working...