Forgot your password?
typodupeerror
IBM IT Hardware

IBM Leaks Details on New Mainframe 185

Posted by ScuttleMonkey
from the it's-all-about-the-benjamips dept.
Mark writes "Big Blue inadvertently revealed details about its new z10 Enterprise Class mainframe set to launch on Feb. 26, as well as details on z/OS v1.10, a new version of the mainframe OS due out in September. 'According to an internal IBM document obtained by SearchDataCenter.com, the z10 Enterprise Class will come in five different models and feature 64-way chips, compared with the 54-way z9 mainframes and earlier 32-way models. In a conference call last month, IBM CFO Mark Loughridge told investors that the z10 would have 50% more capacity, which indicates that it will probably tap out at around 27,000 million instructions per second (MIPS) at the top end, compared with about 18,000 MIPS on the previous z9 Enterprise Class.'"
This discussion has been archived. No new comments can be posted.

IBM Leaks Details on New Mainframe

Comments Filter:
  • five different models and feature 64-way chips
    Is that like 54 cores on a chip? Wowzers, we're gonna need a lot of salsa for this party.
  • Nah (Score:5, Funny)

    by dwalsh (87765) on Friday February 22, 2008 @06:28PM (#22521592)
    I am not buying one till they get that OS up to 3.0 at least.
  • Imagine... (Score:5, Funny)

    by palegray.net (1195047) <philip DOT paradis AT palegray DOT net> on Friday February 22, 2008 @06:33PM (#22521656) Homepage Journal
    Just imagine a beowulf cluster of these. It would make my head explode.
  • by TubeSteak (669689) on Friday February 22, 2008 @06:34PM (#22521676) Journal
    How come they talk about thousands of MIPS instead of just saying GIPS?
    • Because it's decimal, not binary, so (if I'm right) it would be BIPS (Billion)

    • Re: (Score:3, Insightful)

      by Enleth (947766)
      It's probably for the same reason we talk about thousands of kilograms instead of "just" saying "gigagrams". The term "MIPS" is not really an abbreviation anymore, it became a proper word describing a performance unit everyone in the industry is used to.
      • by TheThiefMaster (992038) on Friday February 22, 2008 @06:55PM (#22521950)

        It's probably for the same reason we talk about thousands of kilograms instead of "just" saying "gigagrams". The term "MIPS" is not really an abbreviation anymore, it became a proper word describing a performance unit everyone in the industry is used to.
        Actually "thousands of kilograms" would be "megagrams", but we generally call them "tonnes".
        • by QRDeNameland (873957) on Friday February 22, 2008 @07:04PM (#22522068)

          Actually "thousands of kilograms" would be "megagrams", but we generally call them "tonnes".

          And from there it goes to kilotonnes and megatonnes, then I believe a thousand megatonnes is then commonly called a "shiteload" or, in the US, a "fuckload".

          • by rucs_hack (784150)
            And from there it goes to kilotonnes and megatonnes, then I believe a thousand megatonnes is then commonly called a "shiteload" or, in the US, a "fuckload".

            Actually I believe that when such large measures are used, it generally refers to the means to remove the odd troublesome city or two..
          • by nevali (942731)
            Or "fucktonne", as we often say over here.

            (More commonly, it's "shitload" rather than "shiteload"... at least in the UK).
          • by ozbird (127571) on Friday February 22, 2008 @08:36PM (#22522966)
            I believe a thousand megatonnes is then commonly called a "shiteload" or, in the US, a "fuckload".

            A U.S. fuckload is a thousand megatons. A thousand megatonnes is a metric fuckload.
        • Re: (Score:3, Informative)

          by Enleth (947766)
          Well, not enough coffe today, my bad. Megagrams, sure. However, the phrase "thousands of kilograms" actually IS used in speech instead of "tonnes" for some ranges, something like 1000-20000 kilograms, because it allows to express a precise, yet relatively significant mass in a more natural way when this precision matters (that is, most likely below 20t) - I'd rather say "eight thousand four hundred fifty-four kilograms" than "eight and four hundred fifty-four thousandths of a tonne".

          As an added benefit,
          • I'd rather say "eight thousand four hundred fifty-four kilograms" than "eight and four hundred fifty-four thousandths of a tonne"

            Me too, but I'd be more likely to say "eight point four five four tonnes".

      • Re: (Score:3, Informative)

        by Cheesey (70139)
        It's surprising that people are still using MIPS as a measurement! The (now very old) joke was that MIPS really stood for Meaningless Indication of Processor Speed, on account of the fact that it's highly dependant on how much you can actually do with each instruction, and also which instruction you are measuring. That dates back to the 80s at least, possibly the 70s, and it's why everyone should use representative benchmarks to compare CPUs rather than clock speeds and/or MIPS. The joke even made it into L
    • by avalys (221114) on Friday February 22, 2008 @06:55PM (#22521948)
      MIPS stands for millions of instructions per second, not mega-instructions per second. We'd have to talk about billions of instructions per second, or BIPS, and that sounds lame.

    • by rrohbeck (944847)

      How come they talk about thousands of MIPS instead of just saying GIPS?
      Maybe because "Gips" in German means plaster or cast. That would sound like broken mainframe. Bad marketing.

  • by Doc Ruby (173196) on Friday February 22, 2008 @06:38PM (#22521734) Homepage Journal
    These mainframes use the z6 CPU [wikipedia.org], which is closely related to the POWER6, which is closely related to the PowerPC.

    Is it at all possible to automatically port any nontrivial z6 software to PPC, if it doesn't require the actually different HW of the z6 (or its much higher performance)? Any possibility to run PPC SW on a z6, with some automatic porting for the higher performance?
    • by TheRaven64 (641858) on Friday February 22, 2008 @06:55PM (#22521956) Journal
      The z6 isn't very similar to the POWER6. In the most important aspect with respect to porting software, namely the instruction set, they are completely different. The z6 instruction set is an incremental improvement on an architecture that goes back all the way to the 1960s with System/360 - the longest running architecture to maintain backwards compatibility. The POWER6 architecture is an incremental improvement on an architecture which dates back to the mid '90s and was designed from scratch around a completely different set of ideas.

      The things they share are not visible to the user as they are hidden behind the instruction decoder. You can see some evidence of the fact that IBM are trying to lower costs by sharing a lot of the design between the two lines though from certain new additions to the POWER instruction set, such as hardware support for Binary Coded Decimals (useful in high-throughput financial systems and present in the mainframe line since the 1401 and 700-series, which preceded System/360).

      • by Junta (36770)
        While the frontend may be rooted in the System/360 set, the driving force seems likely to be the same as all the Power6 systems. For example http://www.pseriestech.org/forum/articles/what-is-project-eclipz-112.html [pseriestech.org]

        Note the 'z' in eclipz. They seem to be seeking to consolidate their non-x86 offerings in terms of core component design.
        • by TheRaven64 (641858) on Friday February 22, 2008 @08:31PM (#22522918) Journal
          I've seen that before, but the POWER6 is not the convergence chip. They share a lot, but it's stuff that matters from IBM's perspective (lowers development costs), not from their customers' perspectives (allows them to run the same software). That said, the biggest difference between the POWER6 and the z6 is the instruction decoder (addressing modes are easy to switch - look how many the Core 2 supports). Instruction decoders tend to take a roughly constant number of transistors. One of the big wins for RISC chips was that they could have simpler decoders and dedicate a lot more transistors to execution units. Now, the decoder has gone from being 50%+ of a CISC chip to under 10%.

          It wouldn't surprise me if the POWER7 either implements a superset of the POWER and System Z architectures, or has switchable decoders. Considering the fact that it's already possible to hot-plug CPUs on systems at this level, I can imagine a future IBM line where the hypervisor allows you to not only partition the system, but also decide which chips run in POWER and which in System Z mode dynamically, migrating virtual machines and restarting CPUs as required. That could be very attractive for customers wanting to consolidate mainframe, AIX, and Linux systems.

          One of the design goals of the PowerPC instruction set (a superset of which is implemented by the POWER6) was to easily emulate x86. It would be really interesting if IBM would enhance this emulation support into the hypervisor, allowing customers to run legacy x86 Linux, Solaris or Windows Terminal Server virtual machines on their mainframes.

          By the way, this mainframe is one of the big reasons why IBM are so keen on open source. If you run Linux and (portable) open source software then IBM can sell you a mainframe running Linux VMs when you start to outgrow your current infrastructure. The reason IBM owns so much of the (small, but incredibly lucrative) mainframe market was that in the '60s they pushed the predecessors to this system - System/360. They sold cheap minicomputers and high-end mainframes that ran exactly the same applications (and, with System/370, the mainframe could even run virtual minicomputers). They got people using the cheap minis and then presented them with a clear upgrade path. With open source, they can give people a really long upgrade path starting at commodity hardware and going as far up as they want.

      • by Guy Harris (3803)

        The POWER6 architecture is an incremental improvement on an architecture which dates back to the mid '90s

        Early '90's, as in 1990 [ibm.com].

        You can see some evidence of the fact that IBM are trying to lower costs by sharing a lot of the design between the two lines though from certain new additions to the POWER instruction set, such as hardware support for Binary Coded Decimals (useful in high-throughput financial systems and present in the mainframe line since the 1401 and 700-series, which preceded System/360).

        ..

      • by Drishmung (458368)

        an architecture that goes back all the way to the 1960s with System/360 - the longest running architecture to maintain backwards compatibility

        Since the 1st 360 was introduced in 1964 [wikipedia.org], and the Burroughs B5000 in 1961 [wikipedia.org], you could argue that particular crown belongs to Unisys. If you trace the Large Systems from the B5500 (1964) they are more even.

        • Ah, good point, I keep forgetting the Burroughs Large Systems architecture still exists. The B5000 was a gorgeous architecture - modern systems designers could learn a lot from it - and aspects of it have had a huge impact on the development of several modern programming languages.

          The main difference (from a commercial perspective) is that the S/360 architecture was introduced on a variety of systems. It was designed as a portable architecture from the start, while the B5000 was just such a superb machin

    • Is it at all possible to automatically port any nontrivial z6 software to PPC, if it doesn't require the actually different HW of the z6 (or its much higher performance)?

      Sure, zLinux software can be trivially recompiled for Linux/PPC. But if you're talking about real legacy mainframe code, of course IBM doesn't want people to switch to a cheaper platform. (And I think you mean much higher reliability.)

      Any possibility to run PPC SW on a z6, with some automatic porting for the higher performance?

      Most PPC software is written in C, so you can just recompile it for zLinux if you want to run it slower.

  • n-way (Score:3, Funny)

    by Shotgun (30919) on Friday February 22, 2008 @07:12PM (#22522170)
    Damn. I never ask for enough. There I was, happy for a 3-way.

  • Naming (Score:3, Funny)

    by Tablizer (95088) on Friday February 22, 2008 @07:16PM (#22522220) Homepage Journal
    They should call them something like "mega-servers" instead of "mainframes". They might sell more that way. Hmmm.....iFrame?
    • Re: (Score:3, Informative)

      by PCM2 (4486)
      The name of the product is not as much the stumbling block as is the price. In fact, many (if not most) mainframes are not sold at all, but leased.
  • by Danathar (267989) on Friday February 22, 2008 @07:18PM (#22522244) Journal
    My first two real jobs were as a Computer operator on an old Burroughs system and Sperry/Unisys system. What I find really interesting is how mainframes have really benefited from the same technology that made microcomputers fast. There was a period where clustering PC's (Servers) really was much more cost effective, but as we move into the future the robustness and bulletproof downtime of those old mainframe OS's have been given new life with lightning fast hardware and I/O subsystems.
    • Re: (Score:3, Informative)

      by mgblst (80109)
      Clustered computers are still, and probably always will be, more cost effective than Mainframes. However some applications are more suited towards mainframes, HPC stuff, rather than cluster, HTC stuff.
  • Obligatory (Score:3, Funny)

    by Tablizer (95088) on Friday February 22, 2008 @07:31PM (#22522366) Homepage Journal
    I for one welcome our big and fast Cobol overlords.
  • It looks like IBM leaked details all over their new server. Those detail stains never come out.
  • Big iron Unix servers, and even some larger Wintel servers offer way more competitive pricing and support costs for OLTP systems. However, IBM's support costs for Linux LPARS are about 90% less than the support costs for z/OS.
    I've heard numerous mainframe types tout the performance of mainframes over distributed systems, but I don't buy any of it. The comparisons, as they are in many other areas, are always rigged. When you factor in the cost of "MIPS" (the more you use the more you pay) there is no compar
    • Re: (Score:2, Insightful)

      by warrigal (780670)
      Just look at the customers who buy these machines. Insurance companies and banks will buy six-packs of these new main-frames for their data-centres.

      Current financial situation aside, these people know value when they see it. The mantra "Nobody ever got sacked for buying IBM" doesn't hold up any more. If there was any sort of competition from other platforms these people would buy them.

      In the past manufacturers like Honeywell, Burroughs, NatSemi, Amdahl and so on have built IBM mainframe clones and prosper
      • by peektwice (726616)
        I'd agree that one shouldn't trust huge numbers of transactions to Wintel, because the architecture simply won't handle it. But what about big Unix iron? SunFire and SuperDome systems are able to handle mainframe sized loads (and then some). For large, finely tuned batch jobs, maybe the mainframe has Sun or HP beat, but IMHO, Unix has the best bang for OLTP systems, and are easily as reliable.
        • by ToasterMonkey (467067) on Saturday February 23, 2008 @01:09AM (#22524584) Homepage
          Performance isn't the only issue at hand here. There's also reliability, integration, management, etc. I'm not intimately familiar with IBM mainframe technology, but I've learned enough from people who are to know these are incredibly reliable, and trusted machines. That's why they are used in financial industries, not merely their ability to handle large loads.

          If you were to suggest to to a mainframe guy that he needs to upgrade to a cluster of Unix boxes, you'd get the same look you'd give someone suggesting you should upgrade to a rack of Dell servers. You all think the others are f'ing nuts for different reasons.
  • by jocknerd (29758) on Friday February 22, 2008 @07:47PM (#22522510)
    I guess the IBM rep forgot to mention that a new one was one the way.
  • Let's get them tapes out sooner so I can get a full hour nap in before the day shift comes in!

    Too bad all the new power will likely go toward some new automation to page an admin when his print job abends because it tries to retrieve data from a subsystem during scheduled downtime. Oh well.
    • Translation (Score:3, Informative)

      an ABEND is an 'abnormal end' Which is mainframespeak for when something dies :)
      • by lgw (121541)
        Netware used the same term, which I found humorous when I had moved from Mainframes to Netware - the only thing familiar was the term for failure.
      • Hmm... as a system OPERATOR I already knew that information. But thank you for enlightening the rest of us. It still disturbs my sleep when a job decides to throw a flag because it didn't find data because there is no customer data and I have to write it up. But meh. What am I gonna do? find a real job?
    • I've had a lot of abends because of RACF issues. I remember asking one person to email me their jobcard, and they didn't even know what it was. It seems odd to me to be submitting JCL and not even knowing what a jobcard is.
  • The system goes on-line February 26, 2008. It begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, February 29th. In a panic, they try to pull the plug.
    • Learning the language of large scale of computing is kind of interesting. So far, someone has informed us that an ABEND is an 'abnormal end' Which is mainframespeak for when something dies :)

      And now, ... The system goes on-line February 26, 2008. It begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, February 29th. In a panic, they try to pull the plug

      So ... what is the cute word for "abnormal life" in mainframe world? Even a link to a page of mainframe phrases would be nic
  • by t482 (193197)
    I know the older version supported something like 25 Gbps of I/O. Any idea what this version supports?

  • Who cares? (Score:2, Funny)

    by PPH (736903)

    When clock speeds became so high one could no longer see the bus activity on the cool status lights, mainframes are no longer interesting.

    Bring back the reel-to-reel tape drives while you're at it.

  • Inadvertent (Score:3, Insightful)

    by ShinmaWa (449201) on Saturday February 23, 2008 @01:51AM (#22524760)
    ...my ass.

    That's called marketing son. It comes out in 4 days and they are creating hype for it.

    (NOTE: The inadvertent part was completely fabricated by Slashdot. Not even the article makes this claim.)

"Tell the truth and run." -- Yugoslav proverb

Working...