Slashdot Mirror
IBM's $45 Linux Server (Well, Kinda)
Wedman sent us a snippet from a newsletter from
OpenSourceIT that starts off by saying that IBM will announce a new pricing scheme for Linux
on the S/390 mainframes: Soon they'll cost $125k. For another $20k you can get virtual machine software to run multiple copies of Linux on the same box.
David Boyes, a consultant who works with the S/390, managed
to boot 41,500 Linux servers on one mainframe. Although he notes that
you may not
be able to run that many in real life. ;) (if someone can find an actual link for this, please post it) That just cracks me up: I mean, the debate about forking apache to handle requests is one thing, but hell, why not just boot your own OS for each request!
Phone calls to IBM tech support?
Nope. Different beast. Cray machines are designed for high speed calculation. Also (and I'm sure somebody will jump in here to correct me), Cray can only handle a single operating system image. No support for virtual machines. IBM mainframes specialize in I/O, and business logic, not number crunching. And, the Cray cannot handle multi-terabyte databases, which is IBM's bread and butter.
...phil
...phil
"For a list of the ways which technology has failed to improve our quality of life, press 3."
ok so now the Linux boys can get their hands on the M/F. But why, surely a cluster of 41,000 PC's will still be cheaper than buying a huge great MF and the gubbins you need to support it??
AND I'd like to know if anyone in the real world (ie outside of IBM's labs and Universities) actually use this for anything other than street cred points.
MF's are very expensive to run and even clocking up a few mins of CPU can cost the business huge amounts....
Just wondering why bother? apart from the obvious "because its there"
And when you have to bring the mainframe down due to some moron cutting power mains or some other reason, all 41,000 customers lose service.
I miss IBM mainframes, the are so cool to play with...ah...college days.
IBM just recently announced that they will be selling extra CPUs to their current mainframe users for $125k if they only want to run Linux on them.
m l
:)
:)
http://news.cnet.com/news/0-1003-202-2371079.ht
I guess that's quite a bargain, they normally charge twice that much.
Keep in mind that's just for the processor board, you also have to buy the chassis and memory and disk and... oh I'm sure it's a multi million $ bill when you're all done.
What was realy funy was that IBM had a huge planetwide "Linux/390 Install fest" over the last few weaks.
--= Isn't it surprising how badly I spell ?
Got too many customers, won't be able to handle new requests for VM's, or upgrade the QOS for the old ones? That's cool. Just slap in some more processors.
:)
What's that? No, you don't have to reboot.
PC's may make a great solution for individual desktop computing, but these mainframes kinda rule.
The enemies of Democracy are
What better use of 40,000+ linux "machines" than to build a really big Beowulf cluster.
The I suppose it's not terribly efficient to virtualize a million machines just to tie them all back into one big system. Oh well, it would be fun to try.
"Politicians are interested in people. Not that this is always a virtue. Fleas are interested in dogs." P.J. O'Rourke
It's already built into the operating system. It's called IUCV, "Inter User Communications Vehicle." Basically, it's a connection based, lossless communications protocol that directly moves data between address spaces. Very slick.
Linux on a IBM mainframe was posted on slashdot a while back. There are also linux ibm s/390 resources here.
Preventive War is like committing suicide for fear of death. - Otto Von Bismarck
I've seen the debate start over wether this achievement is useless or not, and it's a good question.
;) Part of the fun of technology is playing skunkworks and experimenting and trying out new things. Sure you get useless stuff, but you also get some real gems.
But, useless or not, it's also pretty cool
Besides, I recall that at one point we'd never need hard drives, or we'd never need more than 640 KB of memory . . . I'm sure someone'll find a use for this one way or another.
"The Sage treasures Unity and measures all things by it" - Lao Tzu
postage and handling.
IT doesn't lack redundancy at all.
The s/390 is fully redundant. Fault tolerant memory (we're talking way beyond ECC here....). THe machine can detect memory errors and route around them.
Fault tolerant processors, fault tolerant coprocessors.. *everything* is realtime fault tolerant.
This machine is *designed* to stay up for a decade, with zero downtime. Nil. Nada. This *IS* a mainframe, a real mainframe. Not some little desktop box.
1) Because this is what the mainframe is *designed* to do. It's not meant to run one single user-level OS. IT runs VM, whichi is designed to do just this. It means you can bang right on the hardware, but virtualized, to protect others who are also using it's massive resources.
2) The point of having 41,500 virtual machines is that each VM is *rock stable*. NOt stable as in 'linux is stable' but stable as in solid concrete nuclear bomb shelter stable. THe 41500 is a test number; in reality you would run maybe a few thousand at most. The point? A hosting provider can provide a linux box to *each* customer! full root access and everything.. go nuts they can say!
"I thought that the PC vs. mainframe debate was settled years ago, and the PC won."
Yep, it was--on the desktop. People with serious reliablity and performance needs (like banks and gov't agencies) still use mainframes.
--
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
I believe that the correct way of phrasing this is that "IBM supplies a Linux Server running on mainframe hardware, pricing starts at $125,000 but drops to $45 in high volumes"
However, I suspect that we'll see people offering virtual hosting within an instance, which kind of defeats the purpose, but also allows, say 100 users to be hosted on an instance, which allows for 2million sites to be hosted on one machine.
Sweet JESUS! Somebody get to work on IPv8 right NOW!
Actaully, if you are talking about massive transaction processing, the mainframes and other large systems (like the AS/400) are the best at what they do, and are actually rather cost-efficient. PC-style hardware has fairly awful I/O performance, which is what really matters for the big iron. Processor power is needed more for application serving, etc.. You can do a lot with big pipes. Sun systems have had great I/O capabilities for a while now, even with slower processors than PC equivalents, and AS/400 and S/390 go far beyond that. Not to mention that the reliability numbers on a 'frame are astounding... but when you a have a bunch of processors that do nothing but handle error recovery, it's no wonder why you don't see problems.
Most people might run maybe one or two VMs with linux, since that could provide a simple way to interface with the rest of the machine for some apps, but obviously, the power of the mainframe doesn't lie in its ability to run linux... it just happens to be a fun little add-on.
--
"It's tough to be bilingual when you get hit in the head."
Ah, but those other 19,999 instances are independently configured, so there's a good chance that the exploit won't exist on all/most/more than one of the instances.
Plus, it might not even be linux running on some/most/all of the instances.
What's so cool about this is that IBM is charging a flat rate for customers to install this on their systems, regardless of how much it is used. Previously, all IBM operating software has been billed on a monthly basis (essentially, you "rent" the operating system), on a pricing scale depending on how many processors you had, how many virtual machine partitions you had, etc.
the S/390 runs VM (Virtual Machine).
;)
Actually, it would be closer to say it runs OS/3x0/Linux, which in turn runs MVS/Hercules, (A S/3x0 emulator for Linux) which in turn runs OS/3x0/Linux. Whew, that was a mouthful!
the machine can virtual within virtual within virtual with no real penalty.
Yep. No penalty, save a tiny amount of MVS overhead. But it's not quite VMware in hardware. MVS is required, but the hardware is designed to help MVS out.
Oh, the networking driver does rock.. It amounts to something like a 10,000T Ethernet card, shoved across the bus. Granted, you can only talk to other machines on the virtual network, that is, inside the S/3x0. Add a single firewall/router session bound to and piping out the real Ethernet feature and you're set tho.
.sig: Now legally binding!
This will never get moderated up, but for those who read down here, the S/390 cpu's that will run ONLY linux, not the ibm mainframe Os microcode, will cost 125,000 USD. These machines will still require at least one REGULAR s/390 CPU, which will run you about 350,000 USD by itself. Never mind the rest of the box. Bottom line, you're still looking at 3/4 of a million to run linux on a Mainframe, so don't get worked up because you have an extra 1/4 million to blow.
Soon they'll cost $125k. For another $20k you can get virtual machine software to run multiple copies of Linux on the same box. David Boyes, a consultant who works with the S/390, managed to boot 41,500 Linux servers on one mainframe.
$125,000 + $20,000 = $145,000 (so far, no $45 server)
$145,000 / 41,500 = $3.50 (lot less than $45)
For the ''server'' to cost $45, that would imply running 3222 (.2 repeating) servers. Is this to be the expected number in that case? I couldn't find this anywhere.
--
Never trust anyone over 90000.
I've got to think Taco knows the answer to that, and he's just trolling.
Anyways, that's obviously not an efficient way to do things. One virtual machine per customer is more likely what you would do with this. (Not to limit one per customer - some customers might actually need more than one, but still, that's the basic idea.) Running a buch of virtual linux boxes isn't going to do anything good for performance, of course, but within reason it shouldn't do much bad either, and you have the advantage of customizing each virtual box individually for it's intended use... different customers won't affect each others machines any more than they would if they each had a physical dedicated server.
And it should be very nice from a security standpoint too, you don't expose the real base OS to the outside world at all, all network traffic goes through one of the virtual linux boxes, and an exploit that compromises one doesn't affect the rest.
Someone will doubtless post the obligatory beowulf cluster comment by the time I press the submit button - of course this is silly. You will get better performance by not imposing the overhead of a virtual beowulf cluster and just dedicating the same resources to a single virtual uber-box.
Performancewise, mainframes are a mixed bag - you can't justify the expense if you are mainly concerned with number-crunching tasks, because a real physical beowulf cluster of alpha boxes, for example, will have a lot more bang for the buck there. Mainframes != supercomputers, they are totally different animals. The mainframes strength is IO, however, so this sort of setup could be very cost effective for massive database applications, web server farms, and the like.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
AFAIK it's about avoiding single points of failure - if one thing breaks/crashes, something else takes over.
Kind of like a cluster in a box.
This message is provided under the terms outlined at http://www.bero.org/terms.html
I wonder if IBM needs beta testers (-: I'd re-wire my house if they sent me a demo unit.
You'd probably hae to get rid of that closet in the corner to fit it into your bedroom. This is not the type of machine to fit under your desk (unless you like your desk 8ft off the ground that is :-) ).
That said, you're probably a little too late. IBM has been helping companies set up Linux (Suse 6.4 I believe) on their S/390's during July so I think that the boat has sailed. Still they'll be playing with the apps for a while yet - I know DB2 is about to be used on that platform, which will be interesting. I'm certain there will be more news sooner or later as well.
Cheers,
Toby Haynes.
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
Of course, price is also an object, so that's why PCs usually win. As for comparing an IBM S/390 to a dual proc Alpha server; it's kind of like trying to compare an oil tanker with a waterskiing boat. It's just not fair to either one to compare them.
"He's more machine now than man, twisted and evil."
I know that I would MUCH rather pay for an S/390 than for the equivalent processing power in PC boxen. I also know that I would MUCH rather administer 1 of these than hundreds of those. That and my AC/Power can cope much better. I think that all in all, I'd be WAY happier with one of these with a decent stack of software running on it than with a handful of PCs.
Eh...
Damn, if you could do this on an H70, my life would be SOOOO much easier. This is exactly the kind of thing I need at work.
"The axiom 'An honest man has nothing to fear from the police'
Why is it that the proponents of "one nation under God" are so eager to get rid of "liberty and justice for all"?
I have used a uk isp called DSVR who seem to be running very high spec linux boxes with virtual machines running on them.
I'm not quite sure how contained each machine was but every customer was certainly given their own copy of apache etc... and it seemed to work pretty well.
Isn't anyone going to ask if I can imagine a beowulf cluster of these?
2 1337 4 u!
Actually AS/400's don't have to be big - I've got one on my desk in front of me. A model 150, it's the same size as a PC, but in black with a red flash...
Single user AS/400s - For when a Sparc isn't exclusive enough.
When I were your age, all round here were fields...
Okay, can someone tell me what the advantage is between having one kernel running, using all the resources, and multiple kernels sharing them. Is it a limit with Linux's ability to use all those processors/memory etc? Or is there a performance advantage doing it this way?
There aren't going to be any official links yet, this was pre-announced at SHARE (www.share.org) in Boston last week. The official announcement should come later this week.
Along with the new pricing scheme, a new product was announced. Called the Virtual Image Facility (this is what costs $20k), it is basically a stripped down version of the old VM OS for the s/390 with some administrative capabilities thrown in for good measure.
Currently Linux only runs on an s/390 in one of three ways, as the only OS on the mainframe, in an LPAR (logical partition, which you are limited to 15 on any given s/390) or as a guest under VM, which is what allows you to run as many as you want (40,000+). The new VIF will allow you to have the benefits of running linux under VM, while actually running under an LPAR (which most shops seem to do, VM is slowly vanishing)
On a side note, Linux was very well represented at this SHARE, with dozens of sessions specifically targed towards Linux. IBM also stressed it's committment to Linux on several occations. My favorite quote was:
"The Penguin is your friend" - Tom Rosamilia, VP s/390 Software Development, IBM
I've found most mainframers seem very open toward Linux (with reservations, these guys are used to reliability and fault tolerance that blowns Linux out of the water). I suspect the reason for this is that the open source development style harkens back to the days many of them remember when IBM released source code for mainframe products (the MVT and HASP days).
Finkployd
LinuxPlanet has just postet an article about this here.
IBM has just held an installfest which they talk about, and they talk with Peter McCaffrey, System/390 Program Director and it looks like IBM is pretty serious about it.
They also talks about what classes of applications performs well on their mainframe and about possible customers.
I don't know the links offhand, but info about Linux on the s/390 has been posted to /. twice before, a month or so ago.
And the 41000 copies, as indicated, but perhaps not emphasized... the guy said that 41000 copies was theoretical, not practical. YOu would not have enough cycles left to actually *DO* anything with that many going.... it was just a test.
But even a few thousand...
IBM, from what I recall, has a neat internal networking driver so the VMs can talk to each other at extremely high speed, which is cool.
The basic idea is that full virtual linux machines can be deployed in minutes, can be cloned, backed up, all kinds of neat mainframe advantages, all on a machine that reall *IS* designed for 0% downtime. No more racks and racks of linux machines... just a fat mainframe.
For those who don't know (and for those who do, correct me if I'm wrong please), the S/390 runs VM (Virtual Machine). The design is such that VM can virtualize itself multiple times over with minimal loss in speed. WE're talking VMWare at the hardware level here... the machine can virtual within virtual within virtual with no real penalty.
Personal Computer -vs- Mainframe? PC won? I think not.
Mainframes are no longer simply 'fast'. PCs can be really fast too.. so can clusters. If all you want is number crunching... perhaps you don't need a mainframe.
WHere the s/390 will SMASH an alpha to bits is on IO. THis thing can MOVE data like you would not believe.
As for performance.. these things have multiple processors I believe, and can be scaled greatly.
IBm has engineered a great solution here. THis isn't simply 'installing linux on an s/390' this is 'running hundreds or thousands' of virtual, fully-working linux systems on one machine.
So the ISP would provide a full linux box to each customer for their site, to do with as they pleased... completely virtual, but hte customer owuldn't see the difference. THe s/390 will control exactly how much resources are sent to each individual instance of linux, so you can have both a) tiers of service and b) performance GUARANTEES. Combined with bandwidth management, this ROCKS. Oh.. you want a faster machine? WE'll just add more cycles to your VM for more money...
This *IS* sweet, from the ISP angle.
but what's a mainframe running Linux gonna do for you?
Stability and speed. For I/O throughput, nothings going to match big iron. IBM has been cranking these things out since the 60's, and they really do know what they are doing.
A really good use would be teaching an operating systems class. Each student would get a virtual machine to play with. Easy to crash, and easy to start right back up. You get the experience of working with a 'real' machine, but not all of the headaches that come from constant reboots. I've taken an OpSys class on both the 390 and x86, and on both I had virtual machines to play with, and it was a nice change of pace to be able to crash the virtual machine rather than the real one. The mainframe had the distinct advantage of being able to host all of the comp sci classes and only start to slow down near the end of the semester. It would only start to be noticeably slow when both CPUs got up to 90% utilization. Not bad for a machine with only 16 megs of physical RAM.
--
then it comes to be that the soothing light at the end of your tunnel is just a freight train coming your way
then it comes to be that the soothing light at the end of your tunnel is just a freight train coming your way
Advantages:
Disadvantages: If all you need is brute processing power (i.e., you aren't doing any transaction-oriented stuff), then run a single copy; you'll get better mileage.
----
That just cracks me up: I mean, the debate about forking apache to handle requests is one thing, but hell, why not just boot your own OS for each request!
I read the article on booting 40000+ linuxii on one box almost a year ago, so I might be a little sketchy, but if I remember correctly:
The issue is booting an independent copy of the OS for each 'instance' of a server, NOT FOR EACH REQUEST. This means that they could run, maybe 20,000 machines that look independent on one shared machine, each of the 20,000 virtual machines running independently of the others. 20,000 root accounts, 20,000 userbases, 20,000 sets of allocated memory, etc., all running simultaneously, off of the same machine.
Right now, web hosts can offer cheap web hosting (virtual hosting), where each user shares the OS, and with properly set permissions, and limited user functionality, this is relatively secure. This is generally run off of one, or a small group of IPs all pointing at the same machine, and the webserver figures out which 'instance' of itself should return what resource to the requester.
The problem is that, for instance, if I need to do something outside of my user-sandbox, I can't, or I need to have someone else do it.
This whole multiple instances of one OS one one bigass machine, appears to the user as a co-location. They don't have to worry about other users screwing with their stuff. Essentially, my instance of the OS on that machine is the same thing as my own box being hosted at the ISP.
AND, with virtual hosting, some user cracks root, and every account on that machine can be comprimised. With this, someone cracks root on one of the 20,000 instances, and whoever maintains that instance gets screwed, but the other 19,999 users are unaffected.
However, I suspect that we'll see people offering virtual hosting within an instance, which kind of defeats the purpose, but also allows, say 100 users to be hosted on an instance, which allows for 2million sites to be hosted on one machine.
I wonder if IBM needs beta testers (-: I'd re-wire my house if they sent me a demo unit.
One University (speak up if you know) is actually running thousands of seperate linux guests under VM on their s/390 and giving EVERY student their own Linux box to play with.
I'm under the assumption that they have some method of dealing with security in a central way so that everyone isn't running tftp and the dreaded r servers.
Finkployd
Don't fork the OS for each web request--fork one for each customer. Think about it--you buy ONE mainframe and ONE copy of Linux. You can 41,000 customers each with their own "machine". They can do whatever they want with it, including configuring the security themselves. It doesn't matter if they do it wrong, the other 40,999 customers aren't affected (with the possible exception of bandwidth).
--
Linux MAPI Server!
http://www.openone.com/software/MailOne/
(Exchange Migration HOWTO coming soon)
You could even implement a good response system to security break-ins. Any time someone logs in as or su's to root, indicate there is an error and swap to one of the hot spares. So what if the cracker trashes the one he is on. Switch to a backup.
Of course in this situation, you would need 100's of spares if someone is a little persistant.
Go not unto/. for advice, for you will be told both yea and nay (but have nothing to do with the question)
David Boyes, a consultant who works with the S/390, managed to boot 41,500 Linux servers on one mainframe. Although he notes that you may not be able to run that many in real life. ;) (if someone can find an actual link for this, please post it)
The story on NetworkWorldFusion News
The story on Fairfax IT
A reprint of the story from LinuxPlanet