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!
No, actually, if you're using VIF (Virtual Image Facility), you don't need the regular S/390 CPU (which is the same as the Linux one, except for the licensing). You can get into the game for easily less than $250K. Admittedly, it's a small mainframe, but it's into the game. And, running VIF, you can give each of your developers a separate Linux box on the box.
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.
I disagreee. PCs are cheap both in price and quality. The best-quality components are much better, but they still fail enough that with 100, 10000, or 41000 you will have several with problems at any one time.
Even with the best server-grade components, the PCs will be far less reliable. And I would bet that 10000 server-grade PCs would cost as much as one of these mainframes. If you own the mainframe, the cost-per-minute charges don't apply (remember that IBM sells now-a-days), and while you need more-expensive operators, you need fewer of them as they are not swapping hard drives and smoking video cards every 20 minutes.
For a long time I advocated clusters of PCs for any application. Clusters have some great uses, but so does real hardware.
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