Slashdot Mirror

← Back to Stories (view on slashdot.org)

IBM Launches Linux-Only Mainframes

Posted by Soulskill on from the blue-tux dept.

An anonymous reader writes: IBM is introducing two mainframe servers that only run on Linux. It's part of a new initiative from the Linux Foundation called the Open Mainframe Project. "The idea is that those companies participating in this project can work together, and begin building a set of open source tools and technologies for Linux mainframes, while helping one another overcome common development issues in the same manner as all open source projects." IBM's hardware release is accompanied by 250,000 lines of code that they're open sourcing as well. "Ultimately the mainframe mainstays are hoping to attract a new generation of developers to their platform. To help coax new users, IBM will be offering free access to the LinuxOne cloud, a mainframe simulation tool it developed for creating, testing and piloting Linux mainframe applications." Canonical is working with IBM to bring Ubuntu to mainframes.

27 of 157 comments (clear)

  1. In Capitalist America... by wonkey_monkey · · Score: 2, Funny

    IBM is introducing two mainframe servers that only run on Linux.

    In Capitalist America, Linux runs on mainframe servers.

    --
    systemd is Roko's Basilisk.
  2. What about Fibre users? by Anonymous Coward · · Score: 5, Funny

    To help coax new users, IBM will be offering free access to the LinuxOne cloud,

    Is this access just for coax users or is it available via fibre or twisted pair?

    1. Re:What about Fibre users? by silas_moeckel · · Score: 2

      Twinax, IBM's old 5250 cable is all the rage again, running new sas/sata displayport and 10-100ge.

      --
      No sir I dont like it.
  3. Ubuntu?! by Anonymous Coward · · Score: 2, Insightful

    I'm sorry, but Ubuntu on mainframes? Ubuntu is the linux distribution FURTHEST from being appropriate for a mainframe - it's heavily targeted towards desktop users, particularly those with a lower level of expertise (or a lower desire to put work into their OS) than the average linux user. What's more, it's adware/spyware now, which is definitely something I'd hate to have on a mainframe - the last thing you want is your OS transmitting and receiving data at random!

    1. Re:Ubuntu?! by Dareth · · Score: 2

      Ubuntu is a popular Linux distribution among developers. While I personally prefer Debian, there are plenty of people using Ubuntu to do some very interesting work.

      --

      I only look human.
      My mother is a halfling and my dad is an ogre, so that makes me an Ogreling
    2. Re:Ubuntu?! by RabidReindeer · · Score: 2

      If they tasked me with deploying Ubuntu to a mainframe by the end of 2018 I'd quit right now.

      Ubuntu... used by windows users who think it's Linux.

      that said.... It's nice to see the forward thinking here, sure we've got solutions in place but this can lead to some very interesting projects in the next few years.

      Oh come on. I'm mainly a Red Hat person, but, excepting Suse, there's no other Linux distro that's more server-friendly than Ubuntu, whose work in that area goes back a decade.

      Right now, the cutting-edge work on clouds and containers is mainly being done on Red Hat and on Ubuntu. And sometimes the more advanced stuff is coming from Ubuntu. If Ubuntu hadn't also corrupted itself with systemd, I'd be seriously considering converting a couple of RHEL boxes.

  4. No it hasn't by drinkypoo · · Score: 2

    These are mainframes that only run Linux. There is no other supported operating system.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    1. Re:No it hasn't by erikscott · · Score: 3, Interesting

      Yup - first thought that ran through my mind: "Oh, they're selling Z Series with crippled Firmware."

      I'm kind of stumped. Linux on a Mainframe is a neat party trick, but it doesn't really make a lot of sense. Modern Z Series hardware is heavily derived from Power. Why not just run Power Linux? Mainframe I/O design is intentionally about as un-PDP-like as possible, so it's a bad match for Unix, Linux, or even Windows for that matter (NT ran on MIPS, so it theoretically could be ported to S/390). Mainframes get their performance by pushing computation into the channel controllers, and while you could do something like that in Linux, are any of your applications ready to treat your database like a device driver? Because that's what you'll have to do. And, incidentally, it's why every attempt from AIX/370 to Linux on Z Series has required virtualization and a ton of independent kernels to get anything resembling decent performance. And that's where Dell will come in and put thousands of cores in a 42U rack for you... No, IBM's own P Series is a better idea, and their former x86 division (now Lenovo) looks even better.

      Erik

    2. Re:No it hasn't by GCsoftware · · Score: 2

      Yeah, I'm sort of failing to see the point here as well - running your classic z/OS backend stuff and then having a few zIFLs talking to the backend over HiperSockets (IIRC) made sense, but just a big zSeries box with no way to run legacy apps?

      I'm not sure but I guess the market will decide..

    3. Re:No it hasn't by bws111 · · Score: 5, Informative

      Modern Z hardware has nothing to do with POWER. Mainframes do not push computation into channel controllers, whatever gave you that bizarre idea? Treat your database like a device driver? What is that supposed to mean? Linux runs native on zSeries, so virtualization is not necessary (and has not been for more than a decade).

      You seem to know absolutely nothing about mainframes, why are you posting?

    4. Re:No it hasn't by bill_mcgonigle · · Score: 3, Informative

      And that's where Dell will come in and put thousands of cores in a 42U rack for you...

      We're getting to the point where all that matters is how much performance can you get from an assemblage of nodes, and how much does it cost to buy and support it?

      If IBM can provide a lower TCO than Dell with different technology and the "containers" are compatible, many customers will be interested.

      --
      My God, it's Full of Source!
      OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
    5. Re:No it hasn't by Tough+Love · · Score: 3, Insightful

      ...of all the things that Linux isn't designed for, is a mainframe OS, because either the mainframe hardware will have to be changed to support Linux's "view" of the world, or Linux has to get a ton of drivers.

      Now I'm stumped by your comment. Linux has no problem with "a ton of drivers", but the fact is, only a few drivers are actually needed.

      I understand IBM's mainframe division is trying to stay relevant, but they need to focus on getting people to use a mainframe, not trying to make a mainframe act as a PC.

      What makes you think Linux makes a mainframe act as a PC? Linux runs on many disparate architectures, some of which look very little like a PC. Sure, Linux forces every architecture to present a page table abstraction derived historically from intel's model, but is that is mainly because that simple model makes sense, and not particularly inefficient for architectures with a different approach to emulate it. Other than that, life in mainframe land is much like any other architecture, especially now that with virualization rampant, everything looks a lot more like a mainframe inside today.

      Things like lockstep CPUs and such, leave that to the hypervisor, and let Linux view it as one CPU in /proc unless there are critical exceptions that need to be passed to the client OS.

      Why would you think it works any differently than that?

      However, what really needs to be done is sell what a mainframe does best, and that is reliability.

      How do you imagine IBM sells mainframes? There was a time when customers had a lot of idle capacity siting around that could be recycled as Linux servers, but now IBM's only compelling argument is reliability. But that argument is a cruncher for some customers.

      What IBM needs to do is have a case for having the hardware be expensive and reduce the number of man-hours needed to be put in to code a solution.

      The mainframe proposition is not about maintenance cost, it is about the business cost of even temporary interruption or failure.

      --
      When all you have is a hammer, every problem starts to look like a thumb.
    6. Re:No it hasn't by erikscott · · Score: 2, Informative

      Z series and power definitely do not share an instruction set, and they have really substantial differences, but that isn't keeping the engineering teams all that separated, if indeed they are at all.

      Quoting Timothy Prickett Morgan from http://www.itjungle.com/tfh/tf... , "And as has been the case in the past, the Power and z processors are designed by a single processing team and are borrowing technologies from each other. This does not, however, mean that IBM is creating a converged processor that can support either Power or z instruction sets." My hazy memory makes me think they're sharing FPU blocks, possibly one of the bus interfaces, and it seems like one of the cache blocks (L3?). Z has plenty of custom hardware - I think it's fair to say it's predominantly custom - the branch predictor would have to be pretty different, and of course power doesn't have a BCD arithmetic unit.

      Point being, if you're going down the Z Series road to run a Unix-like OS, why not just (conceptually) stop early, end up with something like Power, and call it good? Anyway, I'll argue that they're spiritually and economically related, and there's more than a passing family resemblance. Kind of like power and modern ("advanced server") iSeries, though that's getting more into Deliverance territory.

      Meanwhile, channel controllers aren't as dumb as they look. A little wikipedia action here (I know, citing wikipedia, but it's monday and I'm still tired): https://en.wikipedia.org/wiki/... . Turns out the little dickens can do a decent amount of work on its own. I think the wikipedia entry is showing its age... seems like IBM's done a lot more work since this.

      I remember when SASI came out. I looked at the spec and thought "Hey, this is a lot like a channel controller." Then I read some more and decided "No, a channel controller is much smarter. But this isn't bad." SASI became SCSI and everything else flowed downhill from that. At a very real level, Linux is forcing a million dollar fibre channel array to look more or less like an ST506 connected a board from 1984. Wild.

    7. Re:No it hasn't by Guy+Harris · · Score: 2

      Z has plenty of custom hardware - I think it's fair to say it's predominantly custom - the branch predictor would have to be pretty different, and of course power doesn't have a BCD arithmetic unit.

      Actually, it does have IEEE decimal floating-point, as does z/Architecture. z/Architecture has decimal fixed point, but, these days, it might just trap to millicode doing tricks such as excess-6 for carry propagation. (And the PowerPC processors in at least some AS/400 machines added some instructions to assist BCD arithmetic.)

      Anyway, I'll argue that they're spiritually and economically related, and there's more than a passing family resemblance. Kind of like power and modern ("advanced server") iSeries,

      There is no iSeries any more, there's just the IBM Power Systems, which are the successors to both RS/6000^WpSeries^WSystem p and to AS/400^WiSeries^WSystem i; they can run both AIX and IBM i.

      Meanwhile, channel controllers aren't as dumb as they look. A little wikipedia action here (I know, citing wikipedia, but it's monday and I'm still tired): https://en.wikipedia.org/wiki/... . Turns out the little dickens can do a decent amount of work on its own. I think the wikipedia entry is showing its age... seems like IBM's done a lot more work since this.

      Yes, but they're still I/O boxes, not general-purpose computers (well, they might be implemented with z/Architecture or Power ISA processors, but what's exposed to the OS or application programmer is just the ability to run limited channel programs). The z/Architecture Principles of Operation says in "Execution of I/O Operations", in chapter 15 "Basic I/O Functions":

      For subchannels operating in command mode, the channel subsystem can execute seven types of commands: write, read, read backward, control, sense, sense ID, and transfer in channel. Each command, except transfer in channel, initiates a corresponding I/O operation.

      and

      For subchannels operating in transport mode, the channel subsystem can transport six types of com- mands for execution: write, read, control, sense, sense ID, and interrogate. Each command initiates a corresponding device operation.

    8. Re:No it hasn't by erikscott · · Score: 2

      I'm starting to suspect we're in violent agreement here. :-)

      I've physically, with my eyeballs, seen Linux running on some sort of z series a couple years ago. I saw AIX/370 running on some sort of box around 1990-92-ish, so I know it can be done (parenthetically, I was told it shared no code at all with AIX/6000). My entire point with virtualization is not to suggest there's a problem with the mainframe. Whether it makes sense to or not is completely beside the point.

    9. Re:No it hasn't by bws111 · · Score: 2

      Somehow I completely missed the fact that you were talking about the virtualization aspect of it. Sorry about that.

      Even so, I don't think it that big of a deal. Linux on Z has it's own drivers, for DASD, OSA, etc. The virtualization layer only needs to trap the 'start subchannel' instruction and translate the CCWs into the 'real' CCWs then do it's own start subchannel. There is no need for the virtualization layer to emulate things like the channel controllers because that is all invisible to the OS anyway. Z architecture is probably easier to virtualize (I/O wise) because so much of it is invisible to the program. Catch the 'start subchannel' and create an I/O interrupt when done and whatever happens in between is unknown.

  5. And what might this cost? by GGardner · · Score: 3, Interesting

    I notice TFA has no mention of what the hardware will cost, or what IBM will charge for Linux on a mainframe, or even the model numbers of these two mainframes which are Linux only. And MongoDB on an IBM mainframe? Talk about a culture clash.

  6. Re:Really? by Anonymous Coward · · Score: 5, Informative

    Mainframes are nice in that you get hw with 100% uptime. Not 99.99%, but 100%. Electronics getting old? Need replacements? Offline a couple of CPUs, then pull the cards while the machine is running. Insert new ones and bring them up. Repeat, until you've swapped all the CPUs - and the mainframe was running all the time! (Obviously not at 100% capacity, but transactions were processed continuously).

    Memory modules are hot-replaceable in the same manner. So is network, disks & power supplies. All is redundant, all is replaceable without shutting down. You can do such stunts "to some extent" with PC hardware - i.e. you can get a pc-type server board with redundant power. And linux has hot-adding of CPUs already. But mainframes has 50 years of experience with this sort of always-up requirement - so it just works, without snags.

  7. Re:Really? by GCsoftware · · Score: 4, Informative

    Nokia's DX200 series of PSTN switches had fully redundant motherboards, you could literally physically cut the PCI bus and the thing would just keep on rolling, without dropping any calls..

  8. Re:Really? by bws111 · · Score: 2

    The logic is packaged in 'drawers' (up to 4 per system). If one fails it is taken offline and replaced and the image keeps running (at lower capacity of course).

  9. Ubuntu Phone is real. No Red Hat Enterprise phone by raymorris · · Score: 2

    Except for, well the actual facts. Canonical does in fact put Ubuntu on phones. That's actually one of their products, Ubuntu Phone. Red Hat, on the other hand, sells Red Hat ENTERPRISE Linux. They do in fact have a different focus.

    "It's all Linux", one might say. Both do use (different) Linux kernels, just like Android does. There are also differences, such as the focus on new features vs time-tested reliability. Red Hat doesn't get the hottest new stuff the moment that upstream releases a beta. They wait until it's stable and reliable. For mainframes, you probably want stable and reliable.

  10. Re:Really? by Shinobi · · Score: 2

    Mainframes don't have a system bus in the way the PC crowd thinks of it. You can in fact swap out the backplane parts one at a time and maintain system/image uptime/integrity.

  11. Re:Really? by RabidReindeer · · Score: 5, Informative

    Mainframes are not simply overpriced PCs. They're put together internally in quite a different way.

    The original system busses were in the backplane, not in on a "motherboard". That was true even on my very first (S-100) PC, long before IBM got into the personal computer market. The backplane was almost nothing but wiring, with no caps to blow. You'd basically have to set it on fire to render it useless.

  12. Re:It should be more about COBOL? by RabidReindeer · · Score: 2

    So, is the point that if you're spending millions on it, then you will be more careful with the software?
    Do the programmers get to wear a white lab coat?

    Usually more like torn-up jeans and ratty t-shirts. The one who dress fancy are the ones who are least to be trusted.

  13. Re:So what's the point for AIX? by RabidReindeer · · Score: 2

    IBM mainframes were commonly virtual machines. Unlike their predecessors, which had their instructions hard-wired into them, the System/360 and later boxes usually had some sort of "Initial MicroProgram Load" phase that kitted out the machine's NVRAM with the microcode that made them all run the common S/360 instruction set, regardless of underlying hardware, which could be quite radically different, depending on the make and model. Not unlike what Project Hercules provides, but on a much dumber level. In fact, the original floppy disk systems were used as IMPL storage devices on some machines.

  14. Re:So what's the point for AIX? by RabidReindeer · · Score: 2

    Mainframes run zOS, a prosperity OS.

    https://en.wikipedia.org/wiki/Z/OS

    Considering their license fees, it certainly should be a prosperity OS. But I think you meant "proprietary".

    Actually, mainframes have run many different OS's, some proprietary, some custom, a few open-source. Although up until about 1986, the source code for most IBM OS's was freely available.

    Mainframes these days are most likely to run zOS, zVM and/or zVSE. These are the primary IBM licensed OS products. But as I said, other OS's have been implemented as well. I think a number of universities and military installations did so at one time or another.

  15. Ubuntu adware/spyware? by nickweller · · Score: 2

    What spyware, please include citations.

    Scale out with Ubuntu Server

    @Anon: "I'm sorry, but Ubuntu on mainframes? Ubuntu is the linux distribution FURTHEST from being appropriate for a mainframe - it's heavily targeted towards desktop users, particularly those with a lower level of expertise (or a lower desire to put work into their OS) than the average linux user. What's more, it's adware/spyware now, which is definitely something I'd hate to have on a mainframe - the last thing you want is your OS transmitting and receiving data at random!" ref