IBM's Newest Mainframe Is All Linux 251
dcblogs writes "IBM has released a new mainframe server that doesn't include its z/OS operating system. This Enterprise Linux Server line supports Red Hat or Suse. The system is packaged with mainframe management and virtualization tools. The minimum processor configuration uses two specialty mainframe processors designed for Linux. IBM wants to go after large multicore x86 Linux servers and believes the $212,000 entry price can do it."
I guess... (Score:5, Funny)
Re: (Score:2, Informative)
Re:I guess... (Score:5, Funny)
This is all okay with me, as long as they don't use JCL. Ayyy I hated that shit when learning assembler on the 360, but then again they never thought JCL and didn't have any books on it, which is probably why I hated it.
IEBGENR! CORGZ! (runs after IsIIII shouting DASD Acronyms). RACF!!
Ok, ok, it's your lawn.
Re: (Score:2)
Re: (Score:2)
> DD rules :-)
Along with his friend, IEFBR14.
Re: (Score:3, Informative)
Re: (Score:2)
Re: (Score:2)
What's the alternative to JCL?
JCL is like a 60 year old prostitute; it may not be pretty, but it gets the job done flawlessly.
You need documentation (Score:2)
I have all the information I could ever need, right on my desktop [imgur.com]
Re: (Score:3, Informative)
JCL is easy. There are only what, 6 possible statements, and variations on them. And Gary Brown's JCL book has been around since at least the mid-70s iirc. http://www.amazon.com/zOS-JCL-Gary-DeWard-Brown/dp/0471236357 [amazon.com]
IBM 370/VM was ahead of its time... (Score:4, Informative)
I did a lot of work on 370/VM and it was really a brilliant operation system... vastly ahead of its time... it created true virtual machines, with their own virtual hardware and even console/control panel. Within each VM you could run whatever (IBM or non-IBM) operating system you wanted...including another copy of VM to create more VMs... to about 10 levels deep. The implementation was flawless... and each VM was completely isolated. Othere OS have just started to catch up... but most (all?) current OSs don't virtualize the hardware as well.
Back in the hayday of IBM... the system were well documented and incredibly reliable.
I grew to love JCL... alas CICS always sucked.
Re: (Score:2)
Actually, IBM has had a blue penguin for a long time. (Okay, the tux is blue at least)
http://ifup.org/images/tux-genetic.png [ifup.org]
Re:I guess... (Score:5, Funny)
Oh Yeah! (Score:2)
I see those Windows commercials, and I just want to say "Well I'm Ken Thompson, and UNIX was my idea."
Re: (Score:2)
Re: (Score:2)
No need to be sad, here have this nice big blue tie instead.
Year 2009 is the year of... (Score:5, Funny)
Re: (Score:2)
Re: (Score:2)
Nooo!!! We will never have flying cars that way ;( (Score:4, Funny)
( http://www.youtube.com/watch?v=LrFgRAcr0jg [youtube.com] )
x86 Linux Server... (Score:5, Funny)
$212 sounds like a reasonable price for an x86 Linux server, at least as an entry level.
Just one question: What's a "000 entry price"?
Stocking stuffer (Score:5, Funny)
Re: (Score:2)
Re: (Score:2)
Mainframe or Server? (Score:5, Interesting)
There is a big difference between a mainframe and a high-end server. Why would someone buy a mainframe if they didn't need the reliability and special features of a mainframe? Aren't these really the selling points of the Z-series over the P-series, for example? Usually the P-series and I-series systems are also touted for virtualization, and tend to be less expensive. Can someone distinguish the big difference between these lines now? Traditionally, from what I remember, P-series was AIX, I-Series was AS/400, and Z-series was z/OS and other mainframe OS's. Of course, IBM has been offering Linux on all of them for quite awhile now.
Re:Mainframe or Server? (Score:4, Informative)
Correct:
pSeries is the new name for the previous RS/6000 AIX boxes,
iSeries is the new name for the previous AS/400 boxes that ran OS/400,
and zSeries is the new name for the previous mainframe line.
the p and i boxes now run pretty much the same hardware, and both have supported Linux for some time. The iSeries has excellent virtual machine support (called "partitioning" in iSeries parlance) and can run Linux instances natively or on an installed Intel-class processor board that shares system memory and disk (DASD).
As to why you might want to run Linux on mainframe-class hardware, reliability and scalability come to mind.
Re: (Score:3, Interesting)
The fact that so many people make sense of the new names by thinking of them in terms of the old suggests that this was not a great piece of branding. Why did anyone think "we have three distinct product lines with different features and strong brand images within our market, so lets give them all new names, all three of which are nearly the same"
Re: (Score:2)
Re: (Score:2)
It's IBM legacy kit. There are no new customers for AS/400 hardware, so confusing them doesn't matter. Plus, IBM's various divisions are in different cities, and don't play together well. AS/400 = Minnesota, Intel = Raleigh, Mainframe = NY
Re: (Score:2)
Re: (Score:2)
sed 's/\(.\)Series/System \1/g' <<EOF
Correct:
pSeries is the new name for the previous RS/6000 AIX boxes,
iSeries is the new name for the previous AS/400 boxes that ran OS/400,
and zSeries is the new name for the previous mainframe line.
the p and i boxes now run pretty much the same hardware, and both have supported Linux for some time. The iSeries has excellent virtual machine support (called "partitioning" in iSeries parlance) and can run Linux instances natively or on an installed Intel-class processor board that shares system memory and disk (DASD).
As to why you might want to run Linux on mainframe-class hardware, reliability and scalability come to mind.
EOF
Re: (Score:2)
Actually, the name keeps changing. Since the iSeries became the System i, it's changed to the i5, and is now simply called "i".
But most of us who still use and love 'em still call 'em the AS/400. Even the enthusiasts have given up trying to keep up with IBM's marketing division.
Re: (Score:3, Funny)
Re:Mainframe or Server? (Score:5, Insightful)
Re: (Score:2)
Re: (Score:3, Informative)
Re: (Score:2)
You know you've got reliability when its measured in careers.
Modern 'enterprise' PC app: "when did you last reboot?", "we keep rebooting it every 6 hours to free up memory, the guy who implemented it was sacked months ago".
Mainframe app: "when did you last reboot?", "dunno, the guy who did it last retired back in 2002".
Screensaver? (Score:5, Funny)
Re:Screensaver? (Score:5, Funny)
Re: (Score:2, Funny)
I'd rather run Folding@Home, SETI@Home, and GIMPS in the background.
Re: (Score:2)
Re:Screensaver? (Score:5, Funny)
So I was installing SuSE on our new z/OS mainframe (to a virtual machine guest to be specific), and the list of packages being installed was scrolling by: gstreamer.s390
And I'm thinking to myself "who, on God's Green Earth, had the job of porting audio to an EBCDIC based mainframe?" Talk about bizarro world....
But then I thought sure, it may not be used much; but when it does, it could launch 3,840 streams at 130 decibel. It's a Beowulf cluster of Rick Astley in a single box! And THAT is all worth it.
Re: (Score:2, Funny)
Re: (Score:2)
Re: (Score:2)
And why would EBCDIC make audio any harder to implement than on an ASCII or UNICODE based OS/machine?
Because audio files usually come in a human readable, XML based format, of course.
Is your Phil Collins greatest hits collection sounding a bit like a German industrial band? Probably chose the wrong code page for the conversion!
EBCDIC (Score:2, Informative)
The System z hardware is no more EBCDIC than you are. z/Linux uses ASCII for messages, commands, and utilities, just as z/OS uses EBCDIC. The z/Linux choices include ports of RedHat and Suse. The port does not include translating these to EBCDIC.
Now, the old native Unix for z/OS, Unix System Services, *was* an EBCDIC Unix. Nothing ever said Unix had to be based on ASCII. Porting programs to USS was a challenge because too many programs made assumptions about the binary value of characters. Usually the
Re: (Score:2)
Does z/VM do *all* the virtualisation stuff?
I was under the impression that there was some sort of low-level LPAR type partitioning scheme available to?
VM vs LPAR (Score:4, Informative)
With the introduction of XA architecture in the late 80's, IBM moved some of the virtualization technology down into the hardware, they created a new instruction, SIE - Start Interpretive Execution that could tap into this facility. This facility ended up being the heart of both LPARs and VM/XA (which grew into current z/VM). Conceptually the SIE instruction, or the LPAR facility saves the current processor context, and starts a new context. The "guest" system (or the LPAR) now runs in this new context until some condition has been met (e.g. certain timer pops, certain state changes, etc, as defined by the meta-system (z/VM or the base system managing the LPARs). The movement of this function down into hardware was a logical extension of what used to be called hardware VM assists in pre-XA days.
Basically the base hardware provides LPARs (in fact for quite some time IBM mainframes can only run in LPAR mode, even if one has only one system image). LPARs allow sharing of the physical processors, sharing of physical I/O devices, and partitioning of physical memory. With an LPAR you cannot exceed the physical resources available, meaning that you cannot define an LPAR image with more processors then are physically available, or give an LPAR image more memory then is physically available. This is where z/VM comes in.
z/VM provides the ability to virtualize the physical resources. You can define a VM guest with more memory then is physically available, or more processors then are physically available. In addition z/VM can provide virtualized I/O devices, or provide more fine grained partitioning of physical devices (e.g. carving a disk volume into a collecting of smaller volumes in what is called mini-disks -- which are not the same as a disk partition).
Re: (Score:2)
Anyone seriously thinking of plunking the money down for one.
Anyone thinking of spending that much money generally takes some time, or has some flunkies take some time, to learn what they're buying.
Re: (Score:2)
Comment removed (Score:4, Interesting)
Re: (Score:2)
but but but, will it run Windows?
No doubt it will run Windows quite well, from inside a VM.
Re: (Score:2)
Re:That's cool and all... (Score:4, Funny)
Grammar (Score:4, Insightful)
"Its minimum processor configuration are two specialty mainframe processors designed for Linux."
What the fuck kind of grammar is that?
Re: (Score:2)
Poor?
Bad?
Modern?
Re: (Score:2)
Re: (Score:3, Funny)
Re: (Score:2)
Re: (Score:2, Funny)
Re: (Score:2)
Never mind the grammar (I can hardly believe that I'm saying that), I thought that operating systems were designed to work with the processor(s). When did it get to be the other way around?
Re:Grammar (Score:5, Insightful)
I thought that operating systems were designed to work with the processor(s). When did it get to be the other way around?
When it became easier to design a new processor than to design a new OS (and port all apps onto it.)
Re:Grammar (Score:5, Informative)
Never mind the grammar (I can hardly believe that I'm saying that), I thought that operating systems were designed to work with the processor(s). When did it get to be the other way around?
Lots of apps for IBM mainframes are per-processor licensed. This caused a problem for IBM in trying to sell mainframes to run hybrid workloads; the customer would say, "But those extra processors to run Apache on Linux are costing me money in licensing fees on my mainframe apps. It's cheaper for me to buy a smaller mainframe and a bunch of PCs."
So IBM put together a bunch of processors, hardware-identical with normal mainframe processors, but including extra microcode that limits them to running Java/XML (z Application Assist Processor) or Linux (the Integrated Facility for Linux). These units don't count as processors for purposes of licensing mainframe apps, since they can't run mainframe apps.
Re: (Score:2)
Until recently when some wise-ass startup figured out a way to make them general purpose again!
Re: (Score:3, Informative)
I thought that operating systems were designed to work with the processor(s). When did it get to be the other way around?
Around 1970. A lot of minicomputer and mainframe architectures were designed after the initial work was done on the operating system. The OS was designed, then the requirements for the system architecture were extracted from that. Operating systems like MULTICS, VM/360, and VMS had architectures designed to support them, rather than the other way around.
Re: (Score:2)
A typo (typographical error) would be an accidental mis-spelling, such as having started this sentence "A type...", or, perhaps, reversing the relative positions of two words.
This was the apparently deliberate choice of a verb and/or a subject which did not agree with one another in number. An error of grammar and usage, not a slip of the fingers as they fly on the keyboard.
I would want to hack one of those old beats (Score:2)
Question is: Where can I find them? I wonder how MythTV with trans-coding shows and all the rest would run on them. Any ideas on where to find old p- or z- series mainframes?
Re: (Score:2)
I'd rather hack a gibson. Oh yeah baby. You know, one of those super computers they use to do physics and stuff?
Re: (Score:2)
zVM, zOS, TPF/zTPF are better than Linux. (Score:4, Interesting)
There are a number of factors which go into this consideration of Linux on the mainframe. I must admit it was really cool when I first learned of it, having had an MP3000 to myself at an IBM training facilty to learn how to bring up VM/ESA and Linux/390(2001). Then I realized a few things like:
1. Linux cannot take advantage of the advantages of channel-based disk i/o, because it uses Unix i/o approaches which can never be as efficient as the traditional mainframe-based approaches. No one has shown me any evidence that Linux does anything particularly intelligent in its channel program construction and management. Linux assures that IBM can happily sell lots of IFL or general purpose CPUs which are necessary to compensate for this inefficient use of
mainframe resources.
2. Managing workloads under zVM can be great and is extremely well refined, but this requires zVM-specific skills which supposedly no one wants to pay for.
3. For transaction-based work, it's hard to beat TPF/zTPF, but unfortunately that requires some real mainframe skill to implement. And regrettably, zTPF requires Linux and zOS because IBM refuses to convert the programs running on zOS to run on Linux instead. Since TPF/zTPF and zOS both involve onerous monthly licensing charges based on capacity, it's no wonder that TPF/zTPF languish in relative obscurity.
traditional mainframe-based I/O (Score:2, Insightful)
That's interesting, tell us more about the differences between Unix and 'mainframe-based' disk I/O.
Re: (Score:2)
$212,000 entry price (Score:2)
Now none of you can complain that Apple is expensive. Linux/IBM has all of you beat!
Re:An in-house cloud. (Score:5, Funny)
Hmmm... Captive Cloud Computing? Has a nice marketing ring to it. I better hurry and trademark that!
Re: (Score:2, Funny)
Re: (Score:2)
You have no chance against Chinese, with their widely practiced cloud dispersion methods.
Re:An in-house cloud. (Score:5, Interesting)
Now maybe all the companies out there who are thinking of wasting money on cloud computing can just buy one of these, and basically have their own in-house cloud.
Private cloud is flavour of the month it seems. A recent (as in "last month") release from joint venture ACADIA [storagenewsletter.com] (a Cisco+EMC+VMWare+Intel lash-up) shows that packaged private "Cloud" back end server offerings are at least seen as a way people will go.
I think it's smart packaging myself, four-cab VM building block, just add servers and away you go. And since you're just providing a VM environment, you're not limited in your underlying OS choices. Linux is a good way to go there.
~Although the ACADIA system is clearly superior to the IBM offering because (see above link) it can "accelerate customers' ability to increase business agility through greater IT infrastructure flexibility"./~ Gaaaahh!
Re: (Score:2)
If your using linux, why bother with vmware at all?
Just use openvz and get better performance and lower cost. Full virtualization is only worth bothering with if you are stuck running windows as well.
Re:An in-house cloud. (Score:5, Interesting)
If your using linux, why bother with vmware at all?
It's not the environment itself, it's the support stuff. How to manage load balancing a few dozen or even a few hundred servers, what to do with the virtual images you end up with (lots of them. Dedup helps a lot). Server on-boarding VM utilities. Patch management. And do be careful with those DHCP servers, you don't want duplicate address tables.
It's not just running an OS on top of an OS any more. You gotta manage these virtual servers, and that's where people are willing to pay the extra money.
Re: (Score:2)
The load balancing should be pretty easy to script, openvz does live-migrate.The rest seem pretty similar and KSM adds memory deduplication to the kernel so openvz will support that pretty soon.
To me 99% of this stuff seems to be for places that want to hire total dolts and pay somebody else to make it easy for the dolts.
Re:An in-house cloud. (Score:5, Insightful)
To me 99% of this stuff seems to be for places that want to hire total dolts and pay somebody else to make it easy for the dolts.
That's not entirely fair. Sometimes the smartest thing you can do is buy something that already works, not something you have to script to get to work. Boxes you can plug into place. The place for the scripting you're describing is in development. You write a script that works, is flexible and efficient - and then get someone in marketing to put a wrapper around it to convince the buyers it's industry practice. That's what they're doing, anyway.
As good as your solution may be (and I've worn those shoes before myself) you'll be outpointed by the buyers who want it all in a bag taped to the cabinet.
After it's all in place then scripting expertise is needed to keep things working when things need to be changed. Trust me, you don't want a dolt for a sysadmin. I'd rather hire a BOFH [theregister.co.uk] than do that (just remember to keep your own spare key).
Re:An in-house cloud. (Score:4, Interesting)
Re: (Score:2)
Full virtualization is only worth bothering with if you are stuck running windows as well.
HAHAHAHAHAHAHAHAHAHAHAHAHAHHAHAHAHAHH!!!! Oh wait, you were being serious. Oh well.... HAHAHAHAHAHAHAHAHAHAHAHAHAHHAHAHAHAHH!!!!
Re: (Score:2)
Either that or it's well intentioned subversion to get those tricked by the "cloud" salesfolk back on track.
Re: (Score:2)
You should get the IBM DS3200, I hear it's way better.
Re:Can't wait to see the support (Score:5, Interesting)
Um, in 5.5 months, you've had 15 emails? 5.5 months is about 24 weeks or 120 business days; what are you doing only sending email every week and a half?
I had an open software case with HP on an AlphaServer/TruCluster issue that lasted a little over a year. I think I sent over 200 email messages about that case (and there were other people involved as well). We had weekly conference call updates, as well as several meetings with various combinations of HP sales, support, engineers, and managers (many from out of state) in our office. Yeah, it sucked, but part of my job as system administrator is to stay on top of our vendors to make sure they are holding up their end of our support contracts. We aren't any big HP or AlphaServer customer (this was a cluster of two ES40s and represented 2/3 of our total installed base of Alphas, and we didn't have any other HP stuff at all), but we kept on them so they knew they had to deliver.
Re: (Score:2)
No, we are replying within 24 hours. They've sent a tech onsite once. Then they send an e-mail requesting some other piece of info which they "analyze" for 2 or 3 weeks till we poke them again.
Re:Can't wait to see the support (Score:5, Interesting)
I claim BS.
1. You're claiming you wouldn't buy another piece of "IBM server equipment" yet you're complaining about the lowest end disk drive array (really just a shelf) they make.
2. You have no idea what a DS300 is. You claim it's a Fibre SAN device. However, the DS300 is an iSCSI device with RJ45 GigE ports. The DS400 is fibrechannel attached.
You could have had that shelf RMA'd 10x by now. How about picking up the phone? You do have one of those don't you?
Try calling 1-800-IBM-FAST next time.
Re: (Score:2)
How about an IBM Eserver that encountered a warning on boot from the RAID array, not an error and wouldn't boot up until someone pressed a key...
on the console!
so either:
A: I keep my Eserver next to my iPod dock on my desk so I have easy access to press the any key when required.
B: I use Dell and have complete hardware based remote access.
Re:Can't wait to see the support (Score:4, Insightful)
You brought an eServer and are complaining about the lack of remote management!!!
You do realize that with Dell remote management is an optional extra, and there is a range of dirt cheap hardware that comes with nothing more than basic IPMI just like the eSeries does.
You want remote management you need to pay for it.
Re:Can't wait to see the support (Score:5, Insightful)
I call BS. Perhaps incompetence on your part. 15+ emails? In 6 months? Yeah, that is incompetent. A sysadmin who has a serious problem and only sends 15 emails to the supplier in half a year deserves to be on his ass in the street, no severance.
BTW, the problem with your DS300 Fiber SAN is easy to identify. You are trying to use an iSCSI device as a Fiber SAN device. THAT WON'T WORK.
Re: (Score:2)
I call bullshit as the DS300 is the old Adaptec storage array that used old fashioned SCSI with EXP400 expansion shelves. No fibre channel in sight. On that vintage of hardware you would have a DS400 which uses the same EXP400 expansion shelves but presents them as fibre chennel out the back. Still Adaptec rebadged storage.
I would add that we have in general had excellent support from IBM on hardware issues. We where not impressed with the firmware issues surrounding version 7.36.12 and 7.36.15 versions of
Re:Can't wait to see the support (Score:5, Interesting)
IBM sucks. I know, I work for them. If your standard is a help desk staffed by untrained idiots, server support coming from Indians who barely speak english and could care less about anything other than their ticket count, and a consistent effort to make every issue a money issue, then yes, you probably get "excellent" support. I came to IBM in an outsourcing agreement. Before the outsourcing, the IT staff cared about the environment we managed. We wanted things to be done right, and took pride in the work we did. Now, only a skeleton crew of the original employees remain and the rest have been replaced by offshore staff, all at the direction of IBM's upper management determined to put the company and their customers into the poor house while pocketing fat bonuses and exercising stock options funded by the blood of their employees.
Sam P thinks that offshoring is so wonderful, and offers his employees an opportunity to work in "developing" countries for the "prevailing local wage". If he thinks it's so great, then he should move his fat prick ass out of his comfy house(s) and live there himself. No one in the US would miss him. He could take his fudgepacker buddy Bob Moffat with him and they could steal from the locals to pass the time.
Now there will undoubtedly be several who respond that I'm just bitter. They will make comments that I'm just a spoiled American, upset that the cheap labor from other geographies is threatening my lifestyle. And to all of them, from the bottom of my heart, FUCK YOU.
Re: (Score:3, Interesting)
FUCK YOU TOO.
In the 80's, when India was a closed/socialist economy, all the developed countries were banging on our door (some directly, some through IMF, World Bank et al.,) to open up our markets. And the local industrialists were pressuring the government not to open up because they were afraid all the multi-nationals would come and put their protected businesses out of business. Finally, India had to do it out of desperation. Now Indian companies and peoples have learnt the game and found the niches wh
Re:Can't wait to see the support (Score:5, Insightful)
but when they learn the game and beat you with their advantages, it's a bitch. It's a two-way street, dude! Learn to live with it, for you have no choice. If every country went back to their protectionist regimes, the american companies would not survive, for their markets are saturated and there is no growth. They depend on the developing countries' markets for growth.
The problem is, they are not beating the US with advantages.. You see, I'm in the customer service game.. The problem with outsourcing things like customer service, is that customer service is about communicating.. You absolutely can not beat someone in the US at communicating with someone in the US.. I know, I have to listen to all the "thank god's" from customers when they get me instead of our overseas counterparts.. I really don't like hearing it much, because it almost sounds racist.. but it's mostly just venting of frustrations that they have had from previous experiences where they could not communicate well with someone.. The advantages of cheaper labor are lost when a customer has to call multiple times because they can not communicate well with the person trying to help them.. no matter how smart that person may be.. and as some companies have learned, there are customers who will switch to companies who have customer service that they can work with..
Re: (Score:2)
They depend on the developing countries' markets for growth.
No, they depend on developing countries for cheap-ass workers that they treat like slaves whilst reducing relatively expensive workers in the home country at the expense of quality of work.
(not that the developing countries workers couldn't do the work properly, just that the work demanded of them is designed to be cheap and correspondingly focused on cost, not quality)
Re: (Score:2)
Look, IBM has been a great international evil since forever. They made the machines that managed the concentration camps, and the service contract was paid straight to Armonk. IBM's stance has been "that was IBM Germany, it was someone else man!" Anyone who can seriously say that can believe anything.
I went to work for Tivoli right after they had been acquired and had the opportunity to watch a bought-out company lose its soul. When you called Tivoli you used to get someone with a brain on the first try, no
Re: (Score:2)
You sir are an idiot. This is IBM we are talking about here. It could be running any number of proprietary IBM OS's such as z/OS, aix, OS/400, z/VM, z/VSE, z/TPF, or MUSIC/SP.
Probably yes (Score:2)
Re: (Score:2)
It's not really FUD - although it's definitely a marketing ploy !
The only architectural difference between an IFL (Integrated Facility for Linux) and a CP (that's how a CPU that can also run z/OS is called) is how it behaves when the SERVC instruction (0xb220) is issued with the subcode to indicate it wants to determine the hardware configuration (0x00020001))
On an IFL, issuing this specific instruction with this specific parameter will halt the system (more specifically, the system will enter a 'System Che