Slashdot Mirror
The Mainframe Is Dead! Long Live the Mainframe!
HughPickens.com writes The death of the mainframe has been predicted many times over the years but it has prevailed because it has been overhauled time and again. Now Steve Lohr reports that IBM has just released the z13, a new mainframe engineered to cope with the huge volume of data and transactions generated by people using smartphones and tablets. "This is a mainframe for the mobile digital economy," says Tom Rosamilia. "It's a computer for the bow wave of mobile transactions coming our way." IBM claims the z13 mainframe is the first system able to process 2.5 billion transactions a day and has a host of technical improvements over its predecessor, including three times the memory, faster processing and greater data-handling capability. IBM spent $1 billion to develop the z13, and that research generated 500 new patents, including some for encryption intended to improve the security of mobile computing. Much of the new technology is designed for real-time analysis in business. For example, the mainframe system can allow automated fraud prevention while a purchase is being made on a smartphone. Another example would be providing shoppers with personalized offers while they are in a store, by tracking their locations and tapping data on their preferences, mainly from their previous buying patterns at that retailer.
IBM brings out a new mainframe about every three years, and the success of this one is critical to the company's business. Mainframes alone account for only about 3 percent of IBM's sales. But when mainframe-related software, services and storage are included, the business as a whole contributes 25 percent of IBM's revenue and 35 percent of its operating profit. Ronald J. Peri, chief executive of Radixx International was an early advocate in the 1980s of moving off mainframes and onto networks of personal computers. Today Peri is shifting the back-end computing engine in the Radixx data center from a cluster of industry-standard servers to a new IBM mainframe and estimates the total cost of ownership including hardware, software and labor will be 50 percent less with a mainframe. "We kind of rediscovered the mainframe," says Peri.
IBM brings out a new mainframe about every three years, and the success of this one is critical to the company's business. Mainframes alone account for only about 3 percent of IBM's sales. But when mainframe-related software, services and storage are included, the business as a whole contributes 25 percent of IBM's revenue and 35 percent of its operating profit. Ronald J. Peri, chief executive of Radixx International was an early advocate in the 1980s of moving off mainframes and onto networks of personal computers. Today Peri is shifting the back-end computing engine in the Radixx data center from a cluster of industry-standard servers to a new IBM mainframe and estimates the total cost of ownership including hardware, software and labor will be 50 percent less with a mainframe. "We kind of rediscovered the mainframe," says Peri.
There was once a programmer who wrote software for personal computers. "Look at how well off I am here," he said to a mainframe programmer who came to visit. "I have my own operating system and file storage device. I do not have to share my resources with anyone. The software is self-consistent and easy-to-use. Why do you not quit your present job and join me here?"
The mainframe programmer then began to describe his system to his friend, saying, "The mainframe sits like an ancient Sage meditating in the midst of the Data Center. Its disk drives lie end-to- end like a great ocean of machinery. The software is as multifaceted as a diamond, and as convoluted as a primeval jungle. The programs, each unique, move through the system like a swift-flowing river. That is why I am happy where I am."
The personal computer programmer, upon hearing this, fell silent. But the two programmers remained friends until the end of their days.
"First they came for the slanderers and i said nothing."
Are mainframes and PaaS\SaaS really all that different?
Arent PaaS\SaaS just the next step in mainframes?
The IBM pricing really is quite high (there are a ton of licensing fees for the hardware, maintenance, and software). But the systems work reliably. You get a giant system that can run a whole lot of VMs, with fast and reliable interconnects, transparent hardware failover (e.g. CPUs inside most mainframes come in redundant pairs), etc. To get a similar setup on commodity hardware you need some kind of "cloud" orchestration environment, like OpenStack, which can deal with VM management and migration, network storage, communication topology, etc. The advantage of an x86-64/OpenStack cluster solution is that the hardware+licensing costs are loads cheaper, and you don't have IBM levels of vendor lockin. The disadvantage is that it doesn't really work reliably; you're not going to get 5 9s of uptime on any significantly sized OpenStack deployment, and it will require an army of devops people to babysit it. The application complexity also tends to be higher, because failures are handled at the application level rather than at the system level: all your services need to be able to deal with non-transparent failover, split-brain scenarios, etc. Also the I/O interconnects between parts of the system (even if you're on 10GigE) are much worse than mainframe interconnects.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
IBM dude: It might look like a mainframe, but it's a high-capacity, legacy-compatible, fault-tolerant application server.
Me: What's the difference?
IBM dude: About 200 grand.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
For example, the mainframe system can allow automated fraud prevention while a purchase is being made on a smartphone.
Because that's so much different than preventing fraud on a purchase being made from a desktop PC.
No kidding. The only thing I can think of is that the interconnects and processing power of the mainframe allows more heuristics to be run on people's purchasing patterns. Odd pattern = fraud possibility.
Thing is, right now they often consider the individual's purchasing patterns. What about if a whole lot of people start buying from one company? Different pattern to be spotted, can still be fraud.
I don't read AC A human right
.... the more they stay the same. :)
I keep telling my friends that "cloud computing" is not a new concept. We used to call them "dumb terminals." Not a precise analogy of course but close enough for our purposes. You just know that's going to come full circle in another decade or so.
I want peace on earth and goodwill toward man.
We are the United States Government! We don't do that sort of thing.
The death of IBM's mainframes is happening. It was never going to be an overnight thing though. We just replaced our 2 IBM mainframes which cost us just over 10 million each plus licensing and maintenance costs each year with around 2 million of intel based servers. Yes each of those boxes is almost a little mainframe in itself with 80 cores per machine and 4TB of memory, but they run at a fraction of the cost (with more total processing power than the mainframes they replaced) even when providing full cold standby redundancy. There are 3 other places in town that I know that also run mainframes, 1 has 6 of them all of which they have a 10 year plan to phase them out, another has 2 which will be gone by the end of 2016 and the last is the only hold out in town which is waiting to see how our replacement has gone (so far 6 months in and they are happy, another 12 months and the mainframes will be completely turned off).
Look to the future and imagine the internet of things ... yet all those smart devices relay on the big data consuming services of their proprietors. ...
Why are mainframes so hard to replace you ask yourself. Versatility is my answer.
Sure now, how can a machine weighing tonnes be agile and able to adapt to an ever changing world?
The answer is within that ever changing world itself, ask yourself what is a mainframe exactly?
I see mainframes as ultra concentrated bulks of technology, not measured in mips or dry-stones but in simultaneous multi processing threads.
Rather described in technology level instead of architecture and operating system.
On it's own every mainframe is a supercomputer
No I'm not a mainframe operator, though sometimes I wish I was. Once I've had a summer job caring for a GCOS/Bull system in night time.
I enjoyed the smoothness and sheer power of it. It was the start of my carriere in IT.
Bach says it all.
As long as things like Terminal Services and Citrix exist, the "mainframe" as you know it won't ever die.
Buck Feta. You know what to do.
Mainframes are like really big industrial cars where everything is hugely expensive. They're stupid expensive, but far cheaper than trying to do massive amounts of work with thousands of pickup trucks.
It's like the transporter they use to move the space shuttle with rockets and all ready to go:
http://en.wikipedia.org/wiki/C...
It goes 1MPH, which sounds pretty wuss-tastic in car terms, until you consider how much capacity it has at that speed. It would be basically impossible to accomplish the same thing with any number of VW Beetles without spending years taking apart and reassembling everything each time you wanted to attempt a launch.
That's where mainframes make sense - problems which are really massive, but need to run on one computer. Any problem that can be broken down into smaller chunks can be solved much more efficiently with a network of smaller computers.
As the smaller computers continue to get more and more capable and the technology to break down problems and high speed interconnects become more common, the jobs that run better on a mainframe get more rare and networks of servers become more common.
Mainframes do have one cool thing going for them that is not respected on smaller machines - portability. There's code that's been in use for several decades on mainframes running in a stack of emulators. Each new mainframe gets an emulator to make it possible to act just like an an old mainframe. This means the customer needs to run their code on the emulator instead of having to tweak the code to work on the new mainframe. For jobs that justify mainframe costs, downtime is very expensive, so minimizing additional conversion efforts is huge. Also, it's entirely possible that the last person who knew how some mission critical code worked may have died 40+ years ago and business people aren't big proponents of hiring someone to figure out and rewrite legacy stuff.
SAP ate our mainframes.
Skinned. Gutted. The good stuff was ripped out. Sliced. Diced. Fed to SAP one table, one business rule at a time.
The greybeards sit around telling war stories of old as the mainframes fade into obsolecence one LPAR at a time winding down down towards the final CESF.
Sigh.
"We kind of rediscovered the mainframe," says Peri.
Everytime IBM announced a new mainframe line of products they hired an "external" consultant to say exactly the same bullshit. Since 1990, they announced the rediscovery of the mainframe at least five times. IBM is addicted to the mainframe given the large chunk of revenues associated to it. So, they serve us the same marketing bullshit each time.
Achille Talon
Hop!
I found this in the Overview of the Announcement Letter
... the role of the mainframe in the new digital era of IT."
"The name change serves to signal
Us old farts are envious of the new digital mainframes - we were seriously handicapped back then, working on all those old analog mainframes.
It isn't that mainframes are eternal, it's that marketing wonks who write this sort of stuff are allowed to breed...
"Cock Up Your Beaver" does not mean what you think. This sig is intended to clog filters and annoy do-gooders
The difference is the same sort of difference between cellular networks and the internet. The latter is entirely end-to-end; the world-wide-web isn't built on routers, it's built on hosts. By contrast, the former operates with dumb handsets and all the smarts opaquely hidden inside the network; there's a large variety of services run by servers that you the end-user just don't get to see, much less (directly) interact with.
What IBM is using as a marketing spiel here is the appliancification of the end-user apparatus, turning into cellular handset-like "terminal" equipment dependent on the smarts in the network (aka "cloud"), with much less visibility to you as an end-user. Maybe this is inevitable, maybe it isn't. It does mean presenting an "it just works"-face to you the end-user, certainly taking insight away, and effectively taking choice away.
They're far from the only ones to do this. Say, apple and google with their walled gardens are busily doing much the same thing. IBM is just selling this as a tool to build such a garden with, to mobile operators and others who'd like to own a piece of some service providing pie. This while owning the computing space providing equipment the service providers use to make money with, renting it out for a fair premium.
IBM releases the Z9M9Z...
The mainframe people I know, when they rarely refer to transactions, have a slightly different meaning from when windows or unix people do it. The mainframe people more often rever to messages, which is a whole discrete task, which can often require multiple database transactions, some computational passes etc. They usually talk about hundreds of thousands of messages per hour, so if it's 2.5 billion mainframe-style "transactions"(messages), it's pretty damn impressive.
I suppose it's more about integration and optimization of resource allocation around transactions.
Most systems cannot do that efficiently due to multiple layers of abstraction and opacity. For instance, your application server isn't supposed to know about what files and databases are accessed by certain modules/classes and in which ways they're accessed (like append, RW or RO, etc), therefore it couldn't even attempt to improve the overall load by re-arranging requests/transactions in better ways to maximize resource utilization and reduce contention and waiting.
Optimized to solve the problems of 1995. Architecting for the IOT will demand integrated approaches using commodity systems, not re-hashed monolithic 60's architectures. IBM has a new hammer and everything looks like a nail, if you ask them.
Organization? You must be joking..
I had been out of (what I used to know) the mainframe scene for a long while. The zOS was great an everybody loved the mainframe with the exception of the financial department. It was because of the "Monthly License Charge" that some Mainframe models used to have, the software was never licensed to you, it was "rented", so if you didn't pay the MLC you must disconnect the mainframe. Is the MLC over? Does anybody knows?
I think "prevailed" is a bit overstating things. Mainframes have more "held on" despite the march of the killer micros.
http://www.minix3.org/
How workable could it be as a general desktop at this point, like to read email and browse the web? And do some development whether with Eclipse or something else, for C, Java, and JavaScript)?
Does Node.js work on it yet?
http://stackoverflow.com/quest...
"Thanks! I did try getting NodeJS to work in Minix3 but it simply did not work, worked with a couple of guys and there are too many unresolved dependencies and its just a pain... I will try other microkernels and see if I have better luck. Thanks for your reply! -- Purefan Sep 15 '11 at 8:11"
Personally, it seems to me we could have a much simpler OS than something UNIX-y based around Forth and Smalltalk somehow... There seems a lot of clutter and inconsistency of naming things in the UNIX world with various abbreviations (especially including command-line programs and their arguments). But perhaps something like Minix as a microkernel could still form a core for that...
Alan Kay's FONC project was a hopeful step in that direction, but I'm not sure it has really delivered more than some interesting experiments?
http://vpri.org/mailman/listin...
But Alan Kay's heart is in the right place, regardless of recent outcomes. It would have been fun to work with him and maybe become the next Dan Ingalls! :-)
http://www.drdobbs.com/article...
"Kay: Yeah. You want to get those from the objects. You want it to be a mini-operating system, and the people who did the browser mistook it as an application. They flunked Operating Systems 101.
Binstock: How so?
Kay: I mean, look at it: The job of an operating system is to run arbitrary code safely. It's not there to tell you what kind of code you can run. Most operating systems have way too many features. The nice thing about UNIX when it was first done is not just that there were only 20 system commands, but the kernel was only about 1,000 lines of code. This is true of Linux also.
Binstock: Yes.
Kay: One of the ways of looking at it is the reason that WYSIWYG is slowly showing up in the browser is that it's a better way of interacting with the computer than the way they first did it. So of course they're going to reinvent it. I like to say that in the old days, if you reinvented the wheel, you would get your wrist slapped for not reading. But nowadays people are reinventing the flat tire. I'd personally be happy if they reinvented the wheel, because at least we'd be moving forward. If they reinvented what Engelbart, did we'd be way ahead of where we are now. "
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
Big government agencies like SS and IRS with legacy software and large client base?
No matter how reliable and maintainable the box itself is, I wouldn't want my business to lose days or weeks of revenue because the datacenter was swallowed by a sinkhole. Having hundreds of cloud compute instances around the world also helps compensate for network latencies and quickly cut expenses during a business downturn.
I am sure there is a scale at which it makes sense to have dozens of these boxes rather than many thousands of separate instances. Just not sure if volume is enough for IBM to recoup their 1B investment. Good luck!
Is that you?
> swallowed by a sinkhole
Or cratered by an asteroid!
"Mainframe declared dead, film at 11".
And within a year or two, IBM announces that they're shipping more mainframes that year than they've ever sold before.
Datapoint: around 2001 or so, some crazy at IBM, using VM (IBM tech first developed in the seventies), maxed out a good-sized mainframe... running 48,000 *seperarate* instances of Linux, and it ran happy as a clam with "only" 32,000.
How many VMs you got on your server?
*I* have a nice toolbox in my head, with hammers and wrenches and screwdrivers, on how to program on everything from MS DOS to mainframes to Linux. I also know how to admin all but the mainframe, but know something of that. What, Ah say, what do you have to compare, Boy? One ballpeen hammer that only works in Windows?
mark, who prefers to use the right tool for the job
http://www.throwww.com/a/7bn
Also has implications for society. Many of his predictions have been solid for years after writing the essay.
- Nick P
That is why there is GDPS (Geographically Dispersed Parallel Sysplex). Have you ever heard of a bank or credit card company or reservation system losing 'days or weeks' (or even seconds) of data due to a datacenter outage? Do you think that is because by some miracle there has never been a datacenter outage?
Honestly, you guys really need to learn what modern mainframes are and what they do before you go spouting off.
India isn't the place I'd send the compute power. Their infrastructure there sucks and you want a mainframe to work and have adequate support.
India is where you get the cheap labor to work on the mainframe code, but the mainframe is best off in the US, closer to the people buying stuff using the code on it, and in a better infrastructure.
You also don't have to pay mainframes any benefits or give them promotions and they still consistently do their work. Unlike people, who want stuff, and frequently don't do what they have been asked to do.
They also don't vote, so they don't get to vote themselves more benefits.
Who didn't think mainframes were like the epitome of cool back in the 80s/90s? Who doesn't, now, want one of those massive computer-as-art Cray installations, with the comfy couch and the processor coolant trickling across some sculpture that served as eye candy as well as a radiator, and the subtle blinkenlights flashing away, seemingly at random? And then came the Beowulf.
If you were me, you'd be good lookin'. - six string samurai
Hence the point that you should not buy those unless you can afford a dozen. If I only need the power of one of those, I would be better off purchasing less powerful/cheaper systems to distribute worldwide.
Why do you need 'a dozen'? You need 2 for redundancy. And they don't even need to be the same. The largest machine has 446 times the capacity of the smallest. Your primary datacenter is configured for your main workload. Your backup datacenter can be the smallest capacity machine with enough CBU engines to cover the workload of the larger machine. CBU engines cost a fraction of a regular engine. When a disaster happens, you activate the CBU engines and the workload is transferred to the backup datacenter, often with no application interuption at all.
Jealous! We're a little smaller (1 EC12, 3 IFL's, 2 CP's) but I wouldn't want to run virtualized Linux on any other platform, and that's coming off a decade of work in the distributed world. Nothing beats z/VM in my experience, I love it!
You are going to serve your Chinese customers from US datacenters? Hehe. Connectivity between world regions is glitchy and high latency. Amazon, for example, provides 9 regions for its compute instances and they don't spend money for these datacenters just for the heck of it.
Even within a region, you are proposing paying for network and other equipment to handle 100% of peak traffic in each datacenter, while one sits idle most of the time. It would be much cheaper, and provide lower latency, to have 3 datacenters capable of handling 50% of peak traffic and all taking traffic at the same time. If one fails or suffers a network partition, the other two maintain a quorum and continue processing requests. In your scheme, a failover will lose at least some recent changes.
As for your "CBU engines", they are regular hardware crippled by DRM. You are paying for your "cheap" backup mainframe by jacked up prices on your primary mainframe. If both were the same, you would have access to more capacity for the same price.
Mainframes do have one cool thing going for them that is not respected on smaller machines - portability. There's code that's been in use for several decades on mainframes running in a stack of emulators. Each new mainframe gets an emulator to make it possible to act just like an an old mainframe.
Actually, since System/360, each new IBM mainframe got a CPU that executed an instruction set that was a superset of the previous mainframe's instruction set, just as, for example, an 80486 executed an instruction set that was a superset of the 80386's. They did have to provide a mode bit for, say, 24-bit addressing vs. 31-bit addressing, but that's about it - there's also a difference between 64-bit mode and the 32-bit modes (24-bit addressing and 31-bit addressing), but that's true of just about every 64-bit processor out there except for Alpha (which was born 64-bit).
So there's no need, at the hardware layer, for emulation, other than the mode bits, to run older S/360 user-mode ("problem state") code, and certainly no need for a stack of emulators.
S/360 and earlier S/370's had options to emulate older non-S/3x0 processors, such as the IBM 14xx and 70xx systems; I don't know whether they still provide that with any level of hardware/firmware assist or if that's just done in software (which it can be, these days).
There's also backwards compatibility in the OS, but I doubt that involves multiple layers, just the continuing ability to support older APIs and ABIs on newer versions of the OS. That's not unique to IBM's operating systems, although they might be stricter about it than Microsoft or Sun^WOracle or... are.
Client Server turned out to be not so good with proliferation of servers in a company. Virtual Hosts on same hardware solves this somewhat but the model is ugly with redundant code per user and onerous per user fees by MS. The internet programming will mature to have clean server code that can be balanced easily and simple client device to plug into server. Maybe this is what IBM is doing but I see that spun off in the upcoming bankruptcy. Sounds like batch system is making a comeback.
because it travels in front of you, going the same way you are going.