Slashdot Mirror
Mainframe Operators Needed
blueforce writes "Computer World is reporting that there's a shortage of skilled mainframe workers on the horizon. Quote: "Getting IT professionals, especially young ones, interested in learning mainframe work isn't easy." No kidding. While I've never worked on a mainframe, I have worked on AS/400's. 3 words - Mind Numb ing. Perhaps it's time for a more long-term solution to the problem. Interesting nonetheless. Who'da thunk it - a shortage in IT. What's next, COBOL?"
The problem is lack of specialized talent. In neither undergraduate nor graduate school (graduated last year) was a single mainframe course offered. The "old timers" who work on mainframes here are their own special group-- very few people are brought in, and certainly it would be a good idea to change this, since mainframes are years ahead of PCs in terms of hardcore OS technology. If colleges didn't focus so strongly on learning VB and Office, maybe CS degrees would mean as much as they used to...
Wer mit Ungeheuern kämpft, mag zusehn, dass er nicht dabei zum Ungeheuer wird. --Nietzsche
I don't entirely understand why mainframe work should be much more mind-numbing than point-and-click or shell-hopping. Would somebody with AS/400 experience explain what makes administration of the machines completely non-automatable, and thus requiring massive amounts of repetitive input?
--Dan
Could anyone who's actually done/doing this mind numbing work actually give some more details as to exactly what this work is? It's hard to get any appreciation for just how technical one has to be and just how, uh, appreciative of repitition one has to be.
"Please hold my purse while I sit down to go pee".
I work at a Fortune 500 company and I know for a fact that we actually *train* legions of COBOL programmers every year to work on our huge IBM mainframe system. (They start at 35k and go to around 40k upon graduation from the 6-month program. Then they're deployed to various groups within the company.) That thing is *still* the brain of our core business. Whether we like it or not. lol
1) Hire vets. People getting out of the service are a good source for these skills. That was where I got my training.
2) Pay more. Companies have to adjust.
This just happens to be interesting because it is unusual in this job market. It's nice to know I have some skills that might be in demand if my current job goes away.
"Give a woman two glasses of wine and some pad thai, and they'll agree to just about anything." the Sports Guy
While the poster might laugh it up, there are millions of bucks invested in COBOL code, and none of the owners are eager to hear on trashing their systems and rewritting things up.
My other OS is the MCP!
We still run a couple of HP3000s...a 928 and a 918. Before that we had another model, I forget which one. In any case, there is 20+ years of custom in-house COBOL programming invested in those systems. Most of that code is still serving its purpose very well. We have started updating the apps and have done some web development with it, but if it works, why change it? The only reason we have even considered migration is because HP has finally pulled the plug on the 3000 line, not because it couldn't serve its purpose. And hey, who doesn't like a half-obscure OS (MPE/ix) running on a 48MHz machine supporting 200+ users?
Geek used to be a four letter word. Now it's a six-figure one.
... haven't we been meaning to get rid of those Visual Basic people? Here's a way.
Hate me!
"They must have good technical skills and be comfortable dealing with repetitive and mundane tasks."
Sign me up!
I had very high hopes that Y2K would make this 'problem' go away. But NO, they managed to patch all those dinosaurs to keep them running.
:)
Now we're having trouble finding people to toss meat at those same dinosaurs!
What's next, COBOL?
I wouldn't knock it. Which job do you think you stand a better chance of getting: a C++/Java/C# job that you're qualified for with 500 other qualified applicants competing for the same position, or a Cobol job that you're qualified for with 10 other competing qualified applicants. Which do you think will demand a better salary?
Jerk.
The middle mind speaks!
From the article:
[quote]A study last year by Meta Group Inc. in Stamford, Conn., found that 55% of IT workers with mainframe experience are over 50 years old.[/quote]
All your mainframe operators are belong to retirement home. They are on the way to senility. You have no chance to update cobol make your time.
*ITS A JOKE. I PLAN ON WORKING TILL IM 80*
Seriously though, I hope mainframe skills make a comeback, properly infused with the spirit of freedom that UNIX and minicomputers have given us. It's no fair that the old timers get to play with all the kick ass hardware. Just imagine a beowulf... Oh fsck it...
Cheers,
prat
Everyone knows robots are cool, so you'll have no shortage of young computer people clamoring to work on the "Cool" AS/400 robo-admin. Even if all they are working on are the routines that tell the fingers what to type on the console...
"There is more worth loving than we have strength to love." - Brian Jay Stanley
... These are the PERFECT jobs to ship overseas to India for $5 an hour :)
The problem is that no one teaches mainframe operations in schools, you basically need to learn by being dropped into it - and not screwing up everything. Fewer and fewer businesses are willing to invest in promising new talent to learn these legacy systems, but their own mainframe gurus are retiring or dying off - so eventually this corporations will 'bleed out' skill-wise.
And no, the mainframe cannot be replaced by a client-server solution. I listened to this moron chant throughout school - mainframes are not dead. REALITY CHECK - there are just some applications where a mainframe makes more sense. Mainframes can handle enormous amounts of data without having to break it up for a cluster, or without being bogged down with I/O like most client-server type solutions. Mainframes are great when you need to handle databases with tons of information in it - and you need to consistantly dig through it. Most machines cannot handle it, and will buckle. Mainframes almost never buckle, unless you are testing new stuff on them (naughty newbie - that's what a test LPAR is for) or you do funky things to them.
How do you circumcize a hilbilly?(Or any guy from Ohio.)
Kick his sister/mother in the jaw.
Death to Reefer Addicts.
--
The main reason for Mainframe worker attrition is those bloody games that keep dropping into the zone, and those users that keep derezzing the workers. Gaagh!! If only we had a guardian to mend and defend, then we might be able to keep some workers around.
Of course there aren't going to be too many people interested in working on mainframes - the average person hasn't even seen one. Compare that with the near unbiquitous PC, and you can see why a lot more people go for the "traditional" computer jobs - people can get comfortable and experienced with them without spending a lot of money.
Posted by timothy on Wednesday March 26, @05:20PM
from the reliable-transportation-and-3-references dept.
blueforce writes "Computer World is reporting that there's a shortage of skilled mainframe workers on the horizon. Quote: "Getting IT professionals, especially young ones, interested in learning mainframe work isn't easy." No kidding. While I've never worked on a mainframe, I have worked on AS/400's. 3 words - Mind Numb ing. Perhaps it's time for a more long-term solution to the problem. Interesting nonetheless. Who'da thunk it - a shortage in IT. What's next, COBOL?"
I'd love to work on mainframes, bigger the better as far as i'm concerned. trouble is universities here don't teach that stuff ( i guess becuase THEY can't find anyone to teach it) and this doesn't just affect old clunker systems. what about new big iron??? there is always massive computing needs, more is never enough it seems.
If you mod me down, I will become more powerful than you can imagine....
Kinda funny that the article seems to start off by blaming "user friendly" software for the shortage. Is it true that no one tinkers anymore?
Or perhaps the problem is that recent CS degrees are getting as devorced from the actual running of computers as mechanical engeneers are from the running of automobiles? Perhaps the field should split so there is a relatively large group of people with highly technical skills to do most of the labor, and another, smaller, group of the cream of the crop to do the actual computer science. This is instead of calling everyone who takes programming classes in college computer scientists.
Galium Arsenide is the material of the future, and always will be.
Perhaps it's time for a more long-term solution to the problem.
I thought the reason this was an issue was because mainframes were a long-term solution to a problem. They're dinosaurs, sure, but remember that dinosaurs ruled the Earth for millions of years before anything smaller or smarter could come take over. Mainframes are solid, reliable, and house very large amounts of very important data that would have been moved to other systems by now, if it were at all easy to do.
Sure it's easier to get help for newer systems, but it also involves paying for new hardware and software as well as that new help. Why should anyone do that in this economy when they can just pay twice as much for a skilled mainframe administrator and keep their current investment?
Me, I'm surprised that universities aren't training their students in these systems as well as everything else. If I could get AS/400 training as easily as I can get Oracle or Java training, I'd be pulling down twice the money I am now.
Real MCSE's know that real main frames run Windows2000 Server!
Unix is just DOS with funny application names (ga-new what?)
15 years ago, in college, our uni bought a (then new) IBM SP2 machine (only the 32 processor (2x16) if I recall correctly).
:-)
I had to do my thesis about multiprocessor interval-based polynomial factoring on it. I wasted a solid 3 months finding someone who could get the PVM installed. They had NO skilled personel back then. I can't imagine they will find them now.
Mainframe experts are a bit like those iraqi weapons of mass destruction. The dudes in charge claim that tey exist, but no-one has ever seen one
When will I end this grieving ? When will my future begin ?
I don't think so. It is a common mistake to associate Mainframe with COBOL and say "boring". That is an ignorant association. Like Windows users saying Linux is for just for geeks. We use the following languages on our mainframes: COBOL(duh), C, C++, PERL, JAVA, Assembler, REXX and CLIST. There's more. We run Transaction servers, WebServers and application servers. All on a good'oll mainframe. Doing about 45Million business transactions a day.
Anyone remember that Woody Allen movie, "Take the Money and Run"?
This was funny in 1967 and is senseless today, because in 1967, nobody's aunt had a computer. And today, nobody's aunt has a AS/400 sitting around. So the only way to learn is on-the-job, which means there's that chick-an-egg problem of: you can have the job if you have experience, you can get experience if you have the job.With PCs today, you just spend an affordable amount of money, and you can start learning.
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
'ing' is not a word.
Mainframes already are a long term solution. They aren't going anywhere anytime soon.
There's easily more COBOL and FORTRAN code out there than there is Perl or Java.
Systems dont get replaced just because they aren't the 'latest thing'. They get replaced because they dont work anymore. The mainframes work.
Interestingly, I read an article not too long ago about increased interest in mainframes and dumb terminals. Apparently there are those wise enough to realize that the client/server model of computing is full of flaws. When was the last virus to target the Sys/36 or HP-3000?
I don't need no instructions to know how to rock!!!!
Business wants lower costs. Business says
"there's going to be a shortage of X". More labor supplied for X. Cost of Labor of X goes down. Business profits go up.
DO NOT GO INTO MAINFRAME ADMIN !!!!!
The few guys that do know deserve all the money they get. They will be un-employable when the
finally pull the plug on the old iron, so they need to stock up.
People react rationally. The market is a good determiner of labor allocation.
There is no need for special legisilation to promote an industry. There is no need to continually increase the guest worker H1-B quota over and over again.
If business needs a special skill, just raise wages. It's amazing, you'll find the labor you need.
That's only part of the issue. I didn't learn squat about sysadmin tasks in University, because the focus is on teaching you how to think about software development, not how to use a particular tool or platform -- that's what tech schools are for.
A far bigger issue, as was already pointed out, is the mind-numbing tedium of being a mainframe operator. Alas, the same applies to being an operator on any system, as your main job is to swap media for backups, stock the print servers, and act as remote fingers when support staff call in on a page.
Regardless of platform, the only operators I knew who were happy with the job were middle-aged people who were more concerned about job stability than job challenge/fullfillment. Many of them were highly skilled, knew more about the systems than the developers, and would have made good developers. They just didn't want the pressure and insecurity that comes outside the data center.
As to "learning VB and Office", it sounds more like a tech school than a university. I've never heard of VB or Office being considered part of the programming course on a university campus. I have seen it offered as a half-credit course to help out students who have no prior experience with basic office automation tools, but who need the basics in order to be able to prepare and submit their coursework.
Another issue with getting people to consider a career as an operator is that the job stability is a smokescreen. Who wants to take a job for lower pay, that has little or no challenge to it, requires dealing with pissed-off user managers, and is subject to termination whenever someone gets a brain-fart about "saving" by outsourcing?
I do not fail; I succeed at finding out what does not work.
I coded cobol during the Y2K 'crisis'. That was a painful job that caused irreverasable damage on my psyche. To this day I break out in a cold sweat whenever I see a green screen and will probably be in therapy for the rest of my life.
It seems like a losing proposition in the IT business to specialize too much. Now that you don't have a lifetime job, you will have a very narrow range of jobs that you can take. Mainframe maintenence isn't really a growth industry (well, maybe when the current engineers start retiring).
So why would you invest big bucks in a college degree for a career in a shrinking speciality in shrinking field?
Solutions: Ship the jobs to India or import Indians, or just grab some guy off the street and train him (or her).
[off-topic]
Maybe I'm just saying this because of the demoralizing state of the economy: I don't think I'm going to stay in the IT business. I spent a lot of money on a BS and MS degree to get into this field and the IT world looks like its heading the way of Old Detroit. It's not too late to go to medical school.
Personally, I don't think it's a lack of interest. There are plenty of unemployed, ambitious IT people. It's simply that HR departments and recruiters don't even consider someone who is willing to learn and has experience in other IT areas (like client-server, or AS-400 for instance).
Anyone with no experience (but is willing to learn, and has an aptitude to do so) is rejected, and very few companies and/or schools are training anybody on mainframe support.
One or the other must change. It's a vicious cycle - no mainframe support being trained, greater demand for mainframe support personnel, no hiring of any mainframe "trainees", repeat.
Guess you haven't heard of MQ message queueing, TCP/IP sockets, Websphere all on the mainframe, and even a Unix (no not Linux) native environment and filesystem and a JVM as standard parts of IBM's Z/OS.
Your comment about the mainframe is as clueful as if I'd said, "Have PCs made significant progress? Last time I used MS-DOS I had to program in QBasic and dBaseII and I only had a monochrome Hercules graphic card."
This is something I would be interested in: Working Mainframes. I'm in their target range (20 years old), and would love to at least get some experience in the mainframe arena. Trouble is, I can't just pick one up and play around with it like I do with all of the other technologies I know.
I am currently a college student, and there is no course on how to run a mainframe. They will teach you the latest and greatest advances in object oriented programming, but when it comes down to how to work mainframes, I'm in the dark.
So, in short, what can I do to learn? They mention having training programs in the article, but where do I go? And how does one "break in" to the mainframe operator industry?
char sig[120] = "\0"
Why do you consider it mindnumbing? I'm 27, and I've been working with AS/400s for about 4 years. They aren't anymore mindnumbing than running an *nix CLI, or point and clicking all day. On a side note, the AS/400 is quite the machine. I could sit and name all the great things it can do that are better and faster than any Intel system, but it'll still be labeled "mindnumbing" because it doesn't play solitaire.
I went to college for this?...
...since I have never worked on Main Frames and don't have an idea how complex they are. But I think most people like easy interfaces, and that's where they focus, and that's where there is oversupply of labor. Find something hard, and not very popular, and learn it well. You'll earn well, have close to zero competition, and be happy. I learned Novell Netware, and at the begining it was HARD, much harder than Linux. If you didn't know how to do something, there was nowhere to point and click to see if it works. But after a couple of nights playing with it, and asking people who worked with Netware in the past (including my mother) I began to get the hang of it. And I'm happy I did. I have few clients, but they pay well, and since they trust me, *I* will be migrating them to other plataforms if they so choose. If I could, and they pay well, I would be learning to work with mainframes
please excuse my apathy
I work for a large uk bank, and I haven't found this to be the case, whilst I work with the unix security team, I cross train with the main frame guys and gals and here they range from 22 to 60 odd... They're not a bad bunch for a night out either...
:)
I don't agree that working with main frames is boring either, the guys that work on ACF/2 and RACF are extremely highly skilled
One odd thing that ammuses me is the Microsoft like feature of capitalising the dataset if you enter a string containing only capitals... almost makes me expect to see clippy pop up
Tim Brown
Chris, you do realize that there is a significant portion of the slashdot population under 18, right? There's no way in hell such little dopes can understand a business need being more important than the latest and greatest. Hmm, going by your lead-in(replacing YEARS of mainframe development with new applications, oh yeah, a definite cost-saving move), you don't get it either.
Things too reliable? Predictable? Functional? Secure? Just can't find anything to do?
Windowsize it!
Just ten lines of Windows code will have you scrambling for hours to try and figure out what in the heck has gone wrong!
Add twenty lines of Windows code and you've got a month worth of worries on your hands!
And for the truly daring: A mission-critical Windows application!
Kiss the wife and kids goodbye! You'll never eat/sleep/bathe in peace again!
Windowsize it!
(Not approved for those with heart conditions/risk of stroke/high blood pressure/pregnant)
When the Mac first came out, I spent about six months reading technical manuals for IBM's OS370. I wanted to actually work with mainframes, but the people that ran the shops acted like it was some holy grail or something, as though you had no chance of setting FOOT in a data center unless you knew a super-secret magic chant or something. I still think the big iron is fascinating, but I've never been quite motivated to resume my interest (the salaries don't really help, either).
As for C++ programmers - someone made a comment regarding competition among "qualified" c++ programmers. I'd argue that the ability to toss some code into a class so that it compiles with a C++ compiler does NOT a C++ programmer make. If you count only those who know both the language, and how to use it effectively, I'd guess that your competition goes way down.
I've worked at banks my entire professional career. They all call their main systems "mainframes." But, considering they're usually just ass-kicking unix or windows 2000 adv. server boxen, what's the difference between the homemade kind?
Could it be the cash? It certainly appears so, when's the last time you spent $50-100k on a box?
Does lots of SCSI, RAID and redudant power supplies make a mainframe? Or lots of noise? Lord knows you'll find no louder boxen than those beasts.
I've worked on all kinds of unix, from an RS6000, to an AS/400 to BSD, but nevertheless, its all *nix and they're all equipped with the same crontab that everyone else relies on. I make and run scripts all the time, and I don't feel any more leet or cool because I'm working on a "mainframe" rather than a "server."
And the pay doesn't go up for a mainframe sys admin compared to a regular sys admin either, let me tell you.
Mainframes don't require that much babying, that's normally caused by manager ignorance, which is to be expected.
As for COBOL, sure I've worked with some COBOL code, you wouldn't believe how much COBOL is still around in the banking world, but if it ain't broke...
Anyway, my point is is that mainframe is a dated term, now synonymous w/ server.
If I'm mistaken, please, let me know.
Can't you just have to sprinkle some Magic Server Pixie Dust on it? I thought when used regularly, servers would solve their own problems.
I'm programming in C++, completely overworked, and running at roughly 2.5 times the normal workload (i.e. 2.5 programs in the time allocated for just one).
No jobs in IT? Perhaps employers do not want to hire people, but there sure is plenty of work to go around...
They must have good technical skills and be comfortable dealing with repetitive and mundane tasks, said Tucker. "They are usually one or the other," he said.
The computer world has come a long way in terms of automating mundane tasks while they have stood at one point in time. That is the problem, not the lack of people smart enough to type commands and follow instruction, but dumb enough to be content doing the mudane tasks over and over for their whole life. Automate the monkey tasks. It's not rocket science. Wake up and smell the 80's.
Long ago people invented things like print servers and backup schedulers. No reason at all humans should still be doing those tasks. That's what most of the the ops did in the two AS/400 shops I've worked in.
-Pete
Soccer Goal Plans
programs at the Y2K thing? I thought I heard some rumblings about it during all of that.
You just watch.
start:
More cheap labor
more cheap labor
more cheap labor
loop
Sig:
Navy nuke sub lifestyle?
When you compare MVS to UNIX to Linux, and do the math, MVS wins big. Billions of lines of mainframe investment are not going anywhere soon. Billions were spent on making legacy systems Y2k compliant, now that the investment has been made, companies are finding it difficult to call for a re-write.
IBM saw this coming a while back. The 390 mainframes were renamed Enterprise Server (and we all snickered). However, the enterprise server is now running Linux, Websphere, integration services, websites, ASPs, and the legacy systems with incredible stability.
It is difficult to find operators because in many mainframe shops the job consists of running print jobs and contacting support staff when alerts occur. It is no longer a career. It would improve if companies started treating it like a first step. Hire some college students or entry level employees and provide a career path to greater opportunity. Isn't that what we all want?
I work for Kaiser Permanente. We still many Memorex Telex 1191 terminals and IBM 4124 printers in production. The most important application on our PCs is Attachmate Extra, a mainframe emulator. I'm 20 and didn't know a thing about mainframe when I started here 2 years ago. Now they've fired all the mainframe only guys that couldn't learn PC stuff so now knowing how to work on mainframe system is part of our job function. Happily we have an expensive contract with IBM so they repair all the hardware, but trying to figure out why a printer won't print from 1 of its 3 applications can be a real bitch. Kaiser is currently moving to an automated medical record system, which I think will finally start phasing out the mainframe systems. Of course that will probably be another 10 years.
I don't know about any technology-based Universities in the US (or the world for that matter)that offer mainframe classes, but I don't even think my college, or any near me, offers mainframe classes for CS Majors
Actually, my college recently announced that the "C And UNIX" course was optional, instead focusing on Java and theory
MIS/CIS probably have very little opportunites compared to CS/EE type ppl. I suspect that many ppl will be more than happy to apply and more importantly, happy to take the jobs. In fact, that may offer them some unique opportunity for these MISers to move into linux away from MS.
I prefer the "u" in honour as it seems to be missing these days.
Playing BOFH is fun!
You can't judge a book by the way it wears its hair.
$ diff --help
Usage: diff [OPTION]... FILES
Compare files line by line.
-i --ignore-case Ignore case differences in file contents.
--ignore-file-name-case Ignore case when comparing file names.
--no-ignore-file-name-case Consider case when comparing file names.
-E --ignore-tab-expansion Ignore changes due to tab expansion.
-b --ignore-space-change Ignore changes in the amount of white space.
-w --ignore-all-space Ignore all white space.
-B --ignore-blank-lines Ignore changes whose lines are all blank.
-I RE --ignore-matching-lines=RE Ignore changes whose lines all match RE.
--strip-trailing-cr Strip trailing carriage return on input.
-a --text Treat all files as text.
I'm a second-generation programmer. I'm in my mid-30s, and I've done little more than play Startrek on a mainframe terminal... I started out with TRS-80's and followed that track.
The first generation of programmers would be represented by my mother. She started working with computers before there was even such a thing as a "Computer Science" degree -- she has a Master's in mathematics. She was big iron, all the way... when I was a kid, she showed me the washing-machine hard drives and taught me to play the aforementioned Startrek.
She retired just a year or two ago, and she was nervously counting the days. Despite being in the airline reservations industry -- home of some of the biggest iron of all -- her skills and experience were held in less and less esteem as client-server and GUI became the buzzwords. Her biggest fear was that some beancounter would declare her mainframe expertise redundant before the magic date arrived.
On the other hand, at about the time the previous generation was sweating it out, my PC-based experience and VB credentials were all the rage, a ticket to ride the Rapid Application Development gravy train.
So here we are... GUI programmers (thankfully, not myself) frantically searching the want ads, and mainframers in demand. Go figure.
Or to put it another way, plus ça change, plus c'est la même chose! (Yes, I'm a French-speaking Dixie Chicks fan. Call John Ashcroft!)
Stressed? Me? Of course not. Stress is what a rubber band feels before it breaks, silly.
Actually, a local sporting goods chain has been advertising for programmers that are "experts" in Cobol, SQL Server, VB, and IIS to move a legacy system from mainframe to Windows. I taped a 10 foot pole to another 10 foot pole, but still wouldn't touch it.
Disregard anything Computer World has to say. The problem in this industry isn't the fact that there isn't any skill around. I'm sure there are young bucks out there that would love to wrap their heads around a mainframe system even if just for the experience; including myself. The problem is that most human resources departments don't know what the hell they are looking for and/or don't have qualified staff to hire for such positions. The people who get their hands dirty and the skill on the "big machines" are usually apprentices of sorts. People who learn and make their living in the academic community or scientific community. They tend to stay in these communities and never leave; ie, they take care of their own. On the business side of thing there needs to be inroads made in the human resource depts everywhere, even if it means hiring someone technical with a business degree or something of the sort. Or maybe training a technical liason for the purposes of hiring and deeming what actually makes sense because looking in the want ads is like reading the comics.
As for someone highly qualified looking through the want ad's nowadays really allows me to see which companies not to bother with and which to send my resume off to. The shortage has nothing to do with lack of skill.
No, try something more like mainframes running VM with multiple Linux parititions running.
This is how IBM has planned on selling more mainframes. You never hear of new shops going straight to OS/390 or z/OS. It doesn't happen. IBM recognizes that Linux is the future.
Imagine being able to pitch Linux on rock solid zSeries mainframe hardware that can run the equivalent of a couple of hundred Linux servers on one system.
In fact, IBM has a development mainframe with Linux VMs available to certain partners. You sign up online via the web, and after you submit, a job is submitted that creates your Linux VM. You're ready to go within a few minutes.
I hear the Nebuchadnezzar is hiring.
How about we hire some of these eager-to-take-my-job-for-less-money foreigners for the boring jobs, and leave the American jobs for the American citizens?
no comment
If you could get AS/400 training as easily as Oracle/Java training. Then everyone else would too. And the pay scale would adjust accordingly.
"Politicians are interested in people. Not that this is always a virtue. Fleas are interested in dogs." P.J. O'Rourke
Its a lack of opportunities. I for one have some IT experience, and have been passed up over and over for entry level mainframe IT jobs because they want someone who already has experience in it.
I know plenty of out of work IT people who'd be eager to learn mainframe IT if it meant a job, they just aren't willing to teach it.
I've been programming for half my life (since I was 8 y/o) in one language or another, and I'm planning on graduating highschool a full year early. I have experience administrating a multi-user server with users ranging from clueless website admins to fellow developers. I also have experience in automotive mechanics and electronics.
1. How much will a graduate/undergraduate degree affect my eventual wages as a programmer?
2. If I got an electrical engineering degree instead of a computer science degree, would I be able to make more as an embedded technology developer? In the automotive industry?
3. Is college much better than highschool course-wise?
4. Will running various website such as this, that, and the other one help me with admissions? With scholarships?
5. If you woke up and it was X many years ago, would you redo college and grad school, or just go on to a career?
6. Know of any good tech scholarships?
You can't judge a book by the way it wears its hair.
My employer's business lives and dies by our mainframe applications. I have a sneaking suspicion that anyone who has spend millions on hardware and applications and spends thousands if not millions every year on licenses likely NEEDS the I/O of a mainframe. There is no way in Jesus-fucking HELL the V.P. that decided to outsource this to India is going to have a job after the first problem brings an application or god forbid the whole box down. I'm assuming you meant just development, right? Not the actual box and administrators? We can't be that naive, can we?
And I will tell you why. Floor space. Data Center Floor space is very expensive. Mainframes are much smaller then they used to be. You can even get a mainframe on a card and run it on a server that fits under your desk. In any case, even with blades, the need for more servers usually gorws faster then floor space and runs ahead of miniturization. With a mainframe, you can run VM (and not the pretend stuff that VMware sells) and MANY Linux sessions ontop of that. Interserver If you don't like the idea of doign this, run DB/2 as the DB on the mainframe, and then use pSeries or some Linux boxen to do your heavy calculations. Mainframes EXCELL at pumping data. Risc machines are math wizards while CISC machines (Windows or Linux) are very good at being client machines. Why don't you run a mixed environment? It is possible to have job scheduling centralized across multiple platforms including mainframes, UNIX, AS/400 and even Windows 2000/2003 servers. And if your a college, BINGO! That is what you should be doing. This way you can present the sudents with a working example of everything they are being taught. Now I am a sysadmin, but I also do mainframe operations. Mainframes can automate a great many things, but silos only do you so good until they fail and need an operator anyway. Robots can't fix it when a tape decides to break. Of our operators I work with, I have one that is my age (32) and one under 30 (aroudn 26 I think) and one middle age. My boss is in his 50's. We have a good mix. In any case, mainframes are good at what they do. NIX boxes are good at what they do. So are PC's. It just depends. If you actually use the best tool for the best job, you will have a mix of systems.
Gorkman
Red Hat is trying to catch up to XP in terms of usability and performance, but they just keep falling behind. Now they are trying to trick users into "Red Hat Network" so they can just get updates for Red Hat's own mistakes. Anything to try and make a profit I guess. Sure the ABMers will fall for it, but the rest of us know better.
Besides MS server 2003 is coming out and will give us so much more functionally than anything Red Hat can hack together. Oh well....you get what you pay for.
http://saveie6.com/
You must work at the same place I do.
We're in the middle of migrating from an AS/400 office/custom library catalog system to MS Office + MS SQL on W2k.
I've been here five years, and I don't recall the AS/400 system ever going down. Terminal servers are down, Citrix servers are, Excahnge servers are down... three of four times a week.
Sure, I don't have to use Officevision any more, but still, I can work pretty fuckin fast with those old apps. I hope to god they keep payroll on them...
hang brain.
I mean, watching Reboot, looks like the only things that mainframe operators do was play games, and have the guardian beat the operator... Guess we need more operators to play more games, to beat the evil Hexadecimal and MegaByte.....
You want to transfer systems to ones that go up and down like a yo-yo? You'll be sorry! I can't blame you for Officevision though...
"History doesn't repeat itself, but it does rhyme." Mark Twain
We have an HP3000 system here where I work (anyone else have exposure to Datatrak?). It is indeed a pretty reliable OS & HW combination, HP DDS drives not included.
It does feel like a step back in time, though. I'm not the operator of this system, but when I've had to babysit it seems as if there's a lot that could be automated with a better scripting language and some other UNIX-style tools (grep, textuitls, perl).
And as for MPE being unintuitive, well, all a matter of perception. I find :HELLO, :BYE, and :LISTFILE much more intuitive than their *nix counterparts. Maybe not as efficient for typing, but hey, that's what UDCs are for. ;->
In my first job out of University, I was a programmer (not an operator) for the R&D division of a multinational that dealt with mainframes. At 22, not only was I a mainframe systems programmer, but most of the work I was doing wasn't on MVS (the IBM flagship mainfram OS), but VSE, the evil, hunchbacked midget brother of MVS.
Trust me: ugly. Nasty, nasty, nasty. As other people have pointed out, I didn't do any mainframe courses at University. What I did at this job was read a _lot_ of IBM manuals, and attend a bunch of IBM courses.
For those that know the territory, I even went on a JCL (Job Control Language) course.
Basically, for those people used to working and developing in the modern GUI and development tool environment - run in fear. The other people I worked with though pointed out that if you knew this stuff, you would always have a job. Something which this article seems to be higlighting.
I must point out that in hindsight it was very good experience. Being taught to read mainframe dumps, and having to deal with things every day on the bit and byte level was a great foundation for my continued career.
I also bailed from the company after being there for just over one year, wanting to get out of the mainfram environment. And trust me - being an operator is WAAAAAY less interesting than being a systems programmer.
Shouldn't that be NMBUSRMNDing?
Finding God in a Dog
Getting IT professionals, especially young ones, interested in using correct grammar isn't easy.
If the system never goes down, how can upper management ever appreciate your valiant efforts at 3:00 am to restore the essential programs underpinning your whole corporations future? Again?
Windowsize it! Because you haven't suffered enough!
They are simple to use and very fast and stable. The OS is built on IBM's DB2 and can handle some seriously hardcore data processing. I use them everyday and wouldn't hesitate to recommend them to anyone with serious data processing needs.
How appropriate, my company just got bought out and the buyer's running a COBOL based database on IBM mainframes.
They were a little offended when I asked if they were planning on migrating to more modern technologies.
I work as an independent/self-employed consultant, and I've done some work on mainframes recently, mainly integrating older message systems/applications with newer ones.
I find that this is an area where there's less underbidding, and the number of "low life 3 months VB/Java course 'programmers' who really should be washing cars" is much lower than in other areas, like custom application development etc.
And, the organisations which has these mainframes seems to have more mature support and maintainance departments. Of course, most of them are not quite up to speed on the latest acronym to hit the press, but they generally do know their stuff.
Also, the QA seems to be much, much more professionalized, with realistic deadlines and milestones. The mantra seems to be "reliable applications", not "stuff that seems to work, quick!".
I saw someone else in the discussion mention that there was no college training on mainframes. Well, doh. If that's the your issue with working with mainframes or other fields in compsci/it, then you should really consider an alternative career path.
I've been working as a programmer for ~7 years, and I'm updating my knowledge all the time. If you're the kind of person that whines about having to attend a course whenever something "new" (most of the time it is usually old concepts wrapped in new, shiny plastic anyway) or unfamiliar is introduced to you, you should really ask yourself "Shouldn't I really be doing something else?"
Ironically enough:
I have an interview next week at UPS corporate. I would working on their IBM mainframe, working with clients and troubleshooting and such.
I am also a fresmen IT major in college. So maybe the job market is looking good for me, eh?
When dealing with MVS try to forget 95% of what you know of computers and RTFM.
JCL, ISPF, TSO, RACF are a drag but it's prereq to know some (a little) of it.
CICS, DB2, MQSeries are fun.
Again RTFM is absolutely needed because even though it is easy, NOTHING comes natural, MVS is weird for unix people, let alone the rest.
If it still exists and you can lay your hands on it, there used to be a manual, something like Job File Control Block, which contains the structures that JCL is (MACRO ASSEMBLER)ed into. It helps to make sense out of peculiarities such as SYSOUT being a DISPOSITION. In ASM (not in the "higher-level" languages) there is RDJFCB or some such that will read the JCL. I have used JCL to control programs, but almost everything you will ever find uses JCL to supply whatever information is missing from the program's DCBs (as well as supply a JOBNAME and a bit of accounting info).
The problem with this line of work, and the reason so many young'uns don't want to get into it is that it's pretty much dead-end. We're talking about machines and technology that are disappearing. I did my internship with a Cobol-shop but was pretty lucky to move on without much "stigma". Other's I know weren't so lucky. I've known developers, as young and fully capable as I if not more so, have a hard time getting non-Cobol, non-mainframe work afterwards because there seems to be stigma attached. Employers seem to have the perception "Oh, you're a Cobol developer. You won't be able to learn C++/Java/whatever". The only way to move out of Cobol would be to take a cut in pay (to start again at the bottom) or hopefully find an employer that's not so dim. Cobol/mainframe work is not exciting and the desire to move along is gonna be there. But if it's going to be difficult to get out of it then who the hell would want to get into it? The problem occurs because languages and architectures are concrete skills that employers will take note of. Knowledge of design methodology, the "computing sciences", and whatnot (the important skills) can be hard to define and too often get over looked.
Oh, go on, check out my job.
A little OT, but could someone please explain why people are still running mainframes and claim they're so great when you can put an Erlang cluster up on cheapo x86s for much less money and higher reliablity?
I was recently head-hunted by a company seeking MVS and OS/390 C coders for Linux development on big iron. It sure felt good to be talking about signing bonuses again. Happily, once my boss found out about it she opened her cheque book, so I'll be grinding Java for a while longer.
Who woulda thought those mainframe years would ever be useful again? Makes me wanna get my Hercules system up and running...
Hercules is the answer to getting quality time in a Big Iron environment:
...
:^)
http://www.conmicro.cx/hercules/
Almost worthy of its own discussion ! votes ?
Hercules is an open source software implementation of the mainframe System/370 and ESA/390 architectures, in addition to the new 64-bit z/Architecture. Hercules runs under Linux, Windows 98, Windows NT, and Windows 2000.
You may also wish to download the original open source mainframe operating system, IBM OS360
There also are a number of versions of Linux you can run on theIBM Z series Mainframes or the Hercules Emulator
Nothing like getting hot with some Big Iron running in your Pentium in the Study
I've now been unemployed for the past 3 months. I'm no longer young (31) but believe me, Id LOVE the opportunity to work with mainframes even if I was coming in at near entry level and doing "mundane, boring, repetitive tasks". However the truth of it is that the entry level jobs are just not being created for people like me. Almost every advertised position requires X years of experience in X number of very specific areas. There are THOUSANDS of *experienced* IT professionals like-me out there RIGHT NOW, who want roles in IT but cannot get the TRAINING in order to give them the experience for these positions. The problem ISN'T that people need convincing to do this type of work. Its that industry is unwilling to create entry level positions, with a *reasonable* salary and make the long term investment in people by providing the training, to give them the experience, to do these jobs. And the stupid thing is that those of us who are a little older, who have commitments, families etc. are the exact type of people who would LAST at these jobs because job security is much more important to us than the latest technology, or the highest salary. So in closing...anyone know who I apply to?
But, I can't get a job doing this. I can't find anywhere to provide training. I can't get a job, because everyone wants 3 years experience in everything under the sun.
Can someone tell me where I can get training and experience when no one is teaching mainframes and no one is hiring unless one has 2-3 years experience?
There is no "-1 offended" or "-1 you don't agree with me" mod options for a reason.
Nuff said.
Once I was a four stone apology. Now I am two separate gorillas.
The hackers don't like a half-obscure OS (MPE/ix) running on a 48Mhz machine!
There's no r007k1t for it d00d!
I was a Systems Programmer from 21 to 24 (I'm still 24, I changed jobs last summer)
/.) thought it would be.
:)
:)
It was completly different than I (and I suspect most people on
First up, I generally worked with people somewhat older than me (like the article suggests, this isn't a field populated with 20somethings). I considered this a good thing because these people had more experience than any three recent comp sci grads put together. I learned a different perspective on computing that I was not exposed to previously. Namely an obsession with stability, uptime, performance, and attention to detail. When new versions of software came out, we didn't test it real quick an put it into production. We would install it on a test system and abuse the heck out of it for weeks (sometimes months) to ensure it performed up to snuff, was stable, and did not break anything in our environment. Configuration options were painstakenly set after a ton of research.
Because it was a small team that maintained our installation (which was a fairly large IBM s/390 setup) I got to experience a pretty wide range of tasks. I installed various software packages, programmed in s/390 assembler, did some performance testing and tuning, lots of troubleshooting, hardware, and even got to port a very large DCE application server from AIX to OS/390. I didn't get too involved with DBA stuff (that was a seperate group) but I did a lot of operating system related work. Speaking of, upgrading the operating system was such a meticilous production that it makes the most complicated unix install look like a DOS install.
Not to say that it was boring or mindnumbing (there was a little of that but just about every job has that aspect). i found it very challanging and rewarding. Maintaining a website that serves data to a few thousand people is nothing compared to being (at least partially) responsible for the machine that runs the administrative, business and academic record keeping functions of Penn State
Also the people I worked with (for the most part) were very excited about Linux and the open source movement (even before it came to the mainframe). Many of them remember their early days when the mainframe OS and important components was open source (think MVT, MVS 3.8, HASP, etc). People took pride in modifying the code to fit their shop's needs better and improve functionality, fix bugs, all the stuff people do with OSS today. To them the open source movement is a long overdue return to the way things used to be.
As for the learning curve, it is steep. Be prepared to be confronted with a mountain of documentation and accept that you will have to learn a different way of doing things. If you are a person who cannot manage a Linux box wihout X windows installed, or you cannot program without an IDE than it is probably not for you. If you are a somewhat meticulous person who, when learning a new language/program/concept likes to really get into it and learn everything, not just what you need to get by, then consider it.
Personally I loved it and only changed jobs because I was offered a really good job that I couldn't pass up. However I could easily see myself going back to it someday (heck, if the market is that deperate maybe I should start consulting on the side
It is a shame that schools do not teach (if they are even aware of the existance of) mainframes and mainframe technology. They have been quietly running the backbone of most large business and universities for decades now, and I don't see that changing anytime soon.
Finkployd
- Run some java apps on it
... there is a full blown Java2 1.4 VM available (what better platform to run a Java VM on but one that implements the concept of VM at it's core) - Run Linux apps -- Suse, Redhat, & TurboLinux (I think) distros available to run in a logical partition
- Run Apache2 web servers
- Run AIX apps on it
This ain't your father's AS400btw: Did you know that when IBM changed the CPU on the AS400 from a 48bit CISC system to a 64bit RISC system (PPC based) there was almost zero application programming changes required
mm
A teacher at school, who leads the IT learning department, said they where asked to teach mainframes from different IT companies, and they did at one time. And the reason they stopped teaching it was because none of these companies would hiring any of the graduates, because they all where looking for people with at least 3 years of work experience in that IT field.
Reliability of big iron comes at a whopping cost more than "boxes one must administrate". IBM's licensing and support fees are a significant portion of it.
I've worked on systems that were 20 years old and I've seen the hacks done to keep the code in tune with business practices. Simply amazing and scary. Today, programming is often about abstraction and reuse. A lot of work goes into this endeavor. Not back then. I'm of course speaking from the specific, but thats my experience.
I advocating moving to the "long-term" solution asap. The Y2K ruse stole thunder from a real opportunity to jump ship.
It hurts to write RPG, EXEC2 and COBOL.
I started off in this fucked world of computer as an operator on an IBM 3083 in 1985. Watching lines scroll across the screen on a terminal, typing in JCL and NCCF commands, cleaning tape drives and clearing jammed paper drives is not something that is appealing. But it had one major plus over the current IT situation: fixed hours. No huge overtime. A fixed set of tasks and you got to leave at 5PM. No huge headaches. Plain fucking sailing compared to the crap that most of us go through these days with fucked up bosses and ratty coworkers and everyone looking over their shoulders and worrying about their jobs.
Fuck bosses. Fuck capitalism. Fuck globalisation.
I can remember when ComputerWorld was a MainFrame magazine (then again, it was a mainframe world). This is just an article from the old-timers at ComputerWorld remembering the good old days when dinosaurs ruled the earth. Training is no substitute for experience and common sense. My favorite computer operator story is as follows:
The company was having a traing class for new mainframe operators, and the class let out for break. Just a few minutes later, the phones went nuts in the data center because so many problems were happening all of a sudden. Turns out the the trainee was practicing for her operations class by killing active jobs, such as CICS, etc. on the console where no-one was paying attention to her. It took the better part of the day to get everything running correctly again.
Simple companies dont respect elder programmers and it people.. no skilled people aroudn to keep their systems operational..
Don't Tread on OpenSource
I used to run the weekend job schedule on an IBM 370 series mainframe (1970's vintage, although this was in the early 90's.)
It wasn't a bad job; start a job, maybe mount a tape or two, wait around anywhere from 10 minutes to over an hour for the job to finish. Print out reports and distribute them. I would bring SciFi books and do lots of reading, because I had so much free time.
As someone previously said, It's a different world out there. OS/390 is an amazing OS, we had 5 instances of it running at once on the mainframe (2 production and 3 development regions) and then I installed Linux too :-D
Initially I was given a measly 5% CPU for my Linux region....but that's more than enough for Linux to make a mark!
Coming back to the point, the reason for the different world can be summarized in two words - Batch Jobs . So the task of the operators will be to keep staring at the console to look out for requests to load up cartridges (yes, thats the primary backup medium) that a job needs to read stuff off or start some massive printing and computational jobs from time to time. I worked for a large conglomerate, employing more than 30,000 people, so every month the pay-check printing job on huge line printers took about 4 days to complete. Other task included checking DASD usage (mainframe harddrives) to check upcoming shortages , etc.
Even programming for it was fun, I was primarily in charge of the opensystems portion of it, including Domino Go Webserver, O-MVS (unix) and Linux. But I can understand why there is a shortage of manpower. The cool technology does not hit the mainframe world, so you won't get to work on wireless communications, kernel hacking, etc. I am now a Linux systems programmer and I dont wanna go back too, but it was an experience worth getting.
Arrgggh..i have been barfing too long...back to work!
There is no patch for stupidity
Visit my blog
WHen I was in college (less than 6 years ago)I took at class in Cobol from there I got a job as Cobol intern for a large rental car company. I then used that exp. when I graduated to get an intro job as a MVS system programmer. My boss saw the very problem in this article lots of people nearing retirement and was willing to spend lots of time training us and understood that the learning curve is huge. Its tough but you can do it
My exposure to mainframes deals mainly with 70's vintage burroughs systems and several IBM units. What these machines all possessed that NO PC or MAC has is a unbelievable amount of data corruption prevention and data fail safe modes. Several times we actually had equipment (especially the really old stuff) burst into flames!!!! Yet no data was ever lost!!! even when catastrophic system failures occured. Do NOT try this with your PC you will NOT be pleasently surprised!!!
Yes things seem arcaine with these devices, becuase they are!!, but they define mission critical! Designed by business for business. and there are plenty of things we could learn from these old systems!!
Everyone would like to hire folks with years of experience. They put that in the ads, in hopes of snagging such a candidate. But I'm aware of places where these jobs go begging.
Systems jobs call for good problem-solving skills above all. If you have experience with computers, if you are a good learner, and if you have problem-solving skills, apply even if you don't have the experience. If you don't get the job, you might get something else that can help prepare you for the big iron.
When all you have is an axe, everything looks like a grindstone.
Is this the best the trolls can do these days? C'mon! Show some pride in your work!
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
Problem is, what they're offering is well below what it'd take to overcome the pain threshhold and dust off those old college text books.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
In 1985, minicomputers were supposed to make the mainframe obsolete. The DEC VAX and VAX clusters were going to replace all those paleolithic mainframes. Know where DEC is now?
In 1990, PC LANs were going to replace mainframes. Intel 386 and 486 PC's connected with coaxial cable and running LANTastic, Windows for Workgroups, and Novell were going to replace those archaic mainframes. What ever happened to LANTastic? And has Novell taken over the world?
In 1995, client server computing was going to replace the mainframe. GUI front-ends written with PowerBuilder, using back end databases like Ingres and Informix were going to send those mainframes to their demise. PowerBuilder? Is that a sports-nutrition bar?
Anyone see a pattern here.....
If your children ever found out how lame you are, they'd murder you in your sleep
Well, here it is 2003, and where are all the engineering jobs? In fact, where WERE all the engineering jobs? They never appeared!
Mind you, I didn't go into engineering because of that propaganda, but it makes it damned hard to get a job now with so many unemployed engineers.
Indeed. Just for fun, we have a test machine that sits outside our firewalls. It is HIGHLY amusing to look at the logs. Of course you can filter out the automated attacks, script kiddies, etc....but occasionally you get the cracker types attempting to figure out what the hell they've stumbled on. We run very weak passwords on it (this box is mostly just for fun, and we could care less if it did get rooted), and still most people can't figure out how to get a remote prompt since we use only VT-MGR protocol and not telnet. Makes for a few laughs anyway.
Geek used to be a four letter word. Now it's a six-figure one.
How the hell can you expect a mainframe to BSOD when the terminals are green screens?
I do not fail; I succeed at finding out what does not work.
Here's the rub: it doesn't matter what school you go to, what major you declare, what classes you take. There is only attitude.
If you enjoy what you're taking, you'll do better.
If you enjoy stretching and are willing to try new things, you'll do better.
If you remember that you're doing it for you, you'll do better.
If you're only going into something for the money, then chances are you will suck because you probably won't enjoy the classes, you won't learn as much, and you'll get discouraged.
Last thing: sit in front, in the center. Ask lots of (thoughtful) questions. That's an easy way to learn, and you'll find your grades will go up, too.
Yeah, right.
Marist College in Poughkeepsie, NY down the road (I assume) from IBM, has an e-learning course on learning mainframe skills. I'm currently enrolled in the beginner class and there is a crap load of information to go through. It's actually quite an amazing system. For those that are interested you can find more information here. http://www.academic.marist.edu/S390/ I am a UNIX guy, but worked with an old ES/9000 a bit in school and it was a quirky system and I didn't like it as much as the UNIX systems we had, but it was an interesting system. Enjoy, Chris
I believe I am qualified to use a mainframe. I am good with Win98SE and WinXP. I know how to defrag, scandisk, install/uninstall programs, and I even know how to create shortcuts to the desktop. How much harder could a mainframe be lol?
...because the problem is not performance.
The problem is reliability. GNU/Linux clusters are still mostly about performance. Once they get to the level of the VMS clusters, then they will have a reliability solution. Remember, it is not about having three years uptime only; it is about surviving high loads, hardware failures, kernel patching etc.
After we have reliability, there is management. There is a great structure provided by IBM, ISVs and IT old-timers to support these systems. SysAnalysts who really document, security people that really control access, production analysts that really know what is running and what touches which files when to do what.
The fast and loose approach taken by understaffed IT departments supporting GNU/Linux simply will not do. Litmus test: to which of the aforementioned categories do Unix SysAdmins belong? Answer: all of them, as there is not staff enough nor management enough to properly manage systems.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
mind numbing is 2 words, not 3.. im surpised i didnt see that posted yet...
-------
"In times of universal deceit, telling the truth becomes a revolutionary act."
-- George Orwell
... quite some time now.
First, operators who were shunned as too old, out-of-touch, and told that their function could be performed better by a semi-trained chimpanzee. Their lot was targeted back in the late 80's for extinction as less and less dependence on mounting tapes and unjamming remote job printers thrusted them into layoffs and early retirement. And many had their salaries capped and wisely jumped to another face of the company business, many went back to school or trained to become programmers, serving the company by writing and fixing code that ran on the big iron. Most got old and retired and there really wasn't a big recruitment drive for the blossoming career of mainframe computer operator.
In the 1990's, mainframe coders got the crosshair painted on them as their skill, experience, knowledge and extensive depth of how company business worked meant nothing to the new herds of bean counters and unenlightened Dilbert-esque pointy-headed managers who gleefully gazed over the spreadsheet cells in anticipation of the fat bonuses they were getting for eliminating those "cells". "These old tired mainframers are dragging us down. We want to be online and build glittering GUI applications" with a shelf life of months, rather than the stalwart and resilient systems that have been running and still run the good bulk of manufacturing and financial systems. So they trimmed staffs, sent jobs offshore and supplanted U.S. mainframe programmers with imported H1B visa holders.
The difference is that most of the hardware still sits in the U.S., so the demand for operations staff now is acute as offshoring operations may not be as lucrative or convenient or even possible in many instances as it is in the case of programmer staffs.
AZspot
"Makes for a few laughs anyway."
It also makes for a good interview. Crack this box, you have a job.
Depends on what type of hacker you mean. I'm trying to chase down a copy of CLIX, for the InterPro Clipper CPU system I've managed to get.
/. when HP decided to finally kill it.
I have heard of MPE, remember the stories here on
And then, there is the HP 1000 at work, about 2 racks in size... is that the ancestor of your box? Wish I could have it, but I exhausted all my favors just getting them to give me the clipper machine instead of trashing it. And I was able to carry it out... I'd need a Uhaul for the HP.
In Australia, even our universities couldn't afford an ibm (compatible) mainframe to teach. All the big employers who had them had to train in house and on the job the staff to operate them.
But now the jobs to operate those things seem to be invisible. I never see them advertised or meet anyone who currently has a job with the mainframes.
As for mind numbing. I used to make a regular trek with home made choc-chip cookies to the operator's center in order to keep them happy, so if I needed something done urgently, they'd be happy to help out. They used to keep each other amused by playing cricket (like baseball) with paper balls and, um, real bats, or playing practical jokes with "bit buckets" (buckets of chads). Occasionally they would have to put more paper into the printer, I've yet to see that successfully automated, ie that the paper could somehow get from the delivery truck to the paper bin by itself? And much trickier if you're dealing with sprockets.
And the backup tapes. Great reels of ribbon like movie film. If one of those stuffed itself it needed several humans to fix it. And it also took humans to get the things from the tape drives (fridge with eyes) to the storage racks and to manage which tapes went to offsite storage. Cartridge tapes and racks were more automated but every now and again the cartridge robot would break or a programmer would need a cartrige out of the chamber to send elsewhere. Ie the old fashioned sneaker net for data.
I guess these days it would be easier to keep oneself entertained with the internet perhaps. The best operators I knew went on to be systems programmers on the IBMs. I used to butter up those guys as well. Very important. I never understood why various members of the applications programmers (ie cobol hackers) used to look down on the ops and sysprogs, because guess who could completely stuff a application run or rescue it or speed it up?
-- it must be true, it's on the internet.
Another piece of evidence: The Al Jazeera story on slashdot does not appear in the main story, but in yesterdays archives. I try posting to it, and get a "Maximum Comments Exceeded!" error... inspite of this being my first post in a week!!! Read a Salon article on similar attempts at censorship. Spread the word. This is no longer the land of the free. Ashcroft rulez here.
I have decades (yes, plural) of mainframe experience, but I haven't touched one in a few years. Since I've been out of work for a while, I tried dipping my toes back in the water.
Fuck them. Most companies with mainframes have downsized all the competent managers, the ones who knew how much investment was tied up in the big iron. Then they use some young, stupid, "fresh" managers who wonder why they can't just replace a billion dollars (and 25+ years) worth of investment with a couple of windoze machines and some minimum wage operators. Several I talked to said straight out they only wanted recent experience, a gap of a few years to design and build a world spanning telecomms network meant I couldn't be trusted with the machines I've worked around for 20 years. Not that anything much has changed, except for a huge Y2K investment. But the pointy-hairs are so devoid of clue they are killing their companies because they will only pay shit wages to recent graduates. Then they complain about the lack of experienced workers when they won't even hire untrained kids to help round out the teams and take up the reins when the time comes.
One place was a former client, where I had designed and built a complex remote mainframe operations centre. The new management knew nothing of my work, but the few remaining old timers told them to contact me for a similar job. They could only offer me enough money to laughably call an insult. They had a fixed budget, which they had mostly spent on hiring recent graduates at close to minimum wage. Those guys had no experience with big systems, and thus caused a lot of downtime, and cost the company a lot of money. After more than six months, they had burned through most of the budget, and had to scrap the whole project. The new managers wanted me to do the 4 month project in 4 weeks, because that is how much money they had left. When I accepted (hey, a few weeks pay for project I've done successfully five times), they backed off, claiming they would wait for someone with immediate recent experience. Six months later, and 50 million euros in losses to the company, HR fuckheads are still trying to find anyone currently in an identical job. Fuck them.
Ooops, this rant was in no way influenced by the many, many fine Bruges Tripel consumed earlier this evening.
the AC
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
Why can't they use + - / * like everybody else?
Mind Numbing...hmm...lets see...i think thats only two words...
> Mainframes can handle enormous amounts of data
> without having to break it up for a cluster
How do they do that? Your Linux application can keep crunching data too, without breaking it up; except that it would take longer.
> or without being bogged down with I/O like most
> client-server type solutions.
You get bogged down if your application does too much chit-chat. If you design your protocol to minimize that, your I/O will be negligible. The mainframe has nothing to do with it - it's the programmer.
> Mainframes are great when you need to handle
> databases with tons of information in it - and
> you need to consistantly dig through it. Most
> machines cannot handle it, and will buckle.
Bullshit. Again, this is not the mainframe, it is the database engine that runs on it. If you implement the engine correctly, it will run on a PC Linux box just fine too.
> Mainframes almost never buckle, unless you are
> testing new stuff on them (naughty newbie -
> that's what a test LPAR is for) or you do funky
> things to them.
Same is true of Linux. I have never had a kernel crash unless I was playing with some bleeding-edge code that wasn't supposed to be used by sane people.
In other words: the PC Linux box is just as good as the mainframe is your Linux software is just as good as the software you run on the mainframe.
Well, I suppose I might have been 17 years old at some point in my life too, but Mr Structured Programmer ended up unemployed for 4 years while I was getting paid nicely as a systems programmer.
My point is: while it's kind of satisfying to renice others' batch jobs into the abyss and sign them up for onearmedlesbianfetish mailing lists, there comes a point when the tedium of changing tapes and re-stocking printers at 3 in the morning is only worth it if you have a good reason for not being at home.
I'd be wary of any reports of IT shortages. If there's truly a shortage, then companies should offer to train people in the needed skills.
Who would want to pay for their own training in this specialized area, and then find that there's no job? You'd be wasting a year or more of your life and several thousand dollars.
Interesting comparison!
Thought not...
200K tps? Bet that's just in testing - how many million users are you catering for?
The only possible use I can think of for an app that needs to work that fast is something like data capture from physics experiments.
If you're doing that, then flat files are better, but for real transactional stuff, a well indexed relational DB (yes - even M$ SQL Server!) will piss on flat file systems.
I except Btrieve, of course - I've programmed for it in the past, and it was (on Netware) fast as fuck, even on a 386 ;-).
The real reason for using a mainframe is stability - a robust bit of hardware running a robust OS is miles better than a glorified consumer OS running on junk.
Having said that, I run a system based on Windoze NT 4, with a small SAN, that has had around 5 hours downtime in the last 3 years - today was the first time in 38 months that we took the SAN down (the suppliers were amazed).
Keep up with skewl, but remember to do some proper work later in life.
oh brave new world, that has such people in it!
The whitespace is significant. Read it again.
The biggest problem with mainframer's is they are dumb as dirt! They have no motiviation to learn what is outside their little boxes. We have an AS/400 right now with 10 AS/400 programmers. Only about 1 of them has any knowledge outside of yahoo.com when it comes to modern technology. It's down right embarrasing, in fact I can't even claim these people as "developers" cause they are not! Why we have a 400 is even a question I have. We spend hours in meetings trying to figure out how to meet the end user expectations which is a GUI interface. So do you keep programming for the 400? Or retrain? If you retrain, what is the point of having an OVERPRICED mainframe to store 1's and 0's? And yes, it's a mainframe, if you walk like a duck, quake like a duck, have stupid non-computer professionals write RPT then it's a stinking mainframe.
/rant over
and yes, I hold alot of resentment for these peoples. Everytime there is a problem, they are first to throw blame because they can't possibly fathom their amazing 400 would ever fail. Ever seen the data structure designs from a 400 programmer? These people are on CRACK! Reduncy galore..
ok, I could go on for ours
IBM SUCKS!!!!
Mainfraim 3081 thru 390 on VM MVS TSO, rexx, tandem, as/400, and 3270 terminal configuration experience. Anyone Hiring I will work for cheap..Really tired of exchange and email :)
errr....umm...*whooosh* *whoosh* Is this thing on ?
IBM has has VisualAge COBOL (a COBOL IDE) for quite a while.
And then, there is the HP 1000 at work, about 2 racks in size... is that the ancestor of your box?
I'm not sure, but it sounds logical. The 2 boxes we have now aren't any bigger than a standard tower pc. The 3000 model that we had before (purchased around 1989, can't remember model off-hand) was the size of a dishwasher, and that was just the actual cpu, bus, and other main logic boards. The disk drives were housed externally in another box about the same size as the cpu, and the tape drive was also external, and again about the same size. I'm told that the ancestor of that machine was even larger!
Geek used to be a four letter word. Now it's a six-figure one.
I did my time as a third shift operator as a student summer employee at IBM's TJ Watson Research Lab. This was the summer between my freshman and sophomore years of college, so I had only had two semesters of formal computer work, and a year of work on APL and BASIC before that.
... a jet plane landing in the machine room, oh, I mean a full head crash on a 3350 disk drive, the washing machine size disks with removable platter assemblies of many (was it 12) platters, and using the very first generation Winchester heads. ... a minor flood caused by a huricane and a clogged stairwell drain. Nothing like water hitting high voltage power in a transforming closest for some excitment. Though, once we were done trying to defend the machine room with mops, we just sat around until power was restored and then saw who could seek out all of the devices that needed powering-up (it was a HUGE machine room and there was no map) and who got THEIR machine up first.
I spent that summer pulling listings off the line printers (remember them?), cutting graphics output printed on thermal paper that was scattered around the floor by its printer (that was state of the art graphics then), and hanging tapes (primarily for the nightly backups). In my idle time, and there was LOTS of it, I read the manuals on the systems from the company library, and then showed the full time permanent operation staff how to customize their environment and write what we now call shell scripts to automate their environment.
Yup, a college sophomore that was very much wet behind the ears was showing the veterans of many years how to make their jobs easier simply by programming their function keys and writing very short shell scripts. That speaks pretty loudly of the type of people who were on the operations staff. And this was at IBM's front line research lab, and they were very clearly hand-picked from everyone available in the general area which had many very large IBM facilities.
Once in a while we had to deal with an emergency...
The normal staff was three third shift operators, including me. We were running three of what were then the biggest monsters IBM made, and possibly the biggest in the world at that time... three 370 model 168 mainframes, with what I believe was the maximum available CPU, Memory, and I/O subsystems.
It was a very interesting, if sleepless, summer. The insights into who the operators were, and thus who the users of many system level tools are, have lasted a career.
By the way, this was the summer of 1976.
(did someone say they wanted to hear some of that great IBM company folklore I heard that summer? I don't think that non-disclosure for that job prevents me from sharing anything at this late time)
If companies are whining that there is a shortage of AS/400 (or other mainframe) operators out there, then they should blame the mainframe manufacturers for making their products inaccessible, not the young IT people for being uninterested. Believe me, I TRIED to learn about the AS/400, but just finding scraps of information is a chore in itself (aside from superficial marketing crap).
I have worked at a bank for two years now. Before I arrived, I had never even heard of the AS/400. On my first day my manager took me into the computer room and proudly pointed out what looked like a black dishwasher sitting in the middle of the room. "And this," he smirked, "is our AS/400!"
I could bore you with stories about how nobody taught me anything, how I had to figure everything out about the vaunted AS/400 just so I could do my fucking job (and subsequently got scolded for "going where I wasn't allowed") and so on, but I won't.
Instead I'll just say this: whenever I tried to ask my manager or co-workers for an overview of the AS/400, he could only say "oh, it's NOTHING like a PC, it's COMPLETELY different." Me: "okay, but how is it different?" Them: "Oh, it's just different, you wouldn't understand."
I have come to is that my manager doesn't know jack shit about his beloved AS/400. He knows how to "make it work," and even though he claims to have some sort of certification, IBM obviously just gave him scraps. If you can't explain a technology in 3 sentences, then you probably don't understand it at all.
But why does he know jack shit, despite working with an AS/400 for 10 years? It is because IBM has purposely kept him uneducated. Everything is hidden in subscription professional sites and bank-account busting certifications from IBM.
I have searched the web over and over again for information on AS/400 crap. All I can find are IBM's boring information libraries for the "iSeries," mysterious subscription sites for AS/400 "professionals" and this page. Try finding a book on AS/400 online or at your bookstore. They suck. If you want to wade through the IBM manuals online, be my guest. My suggestion is you do it just before bedtime.
Ok, MAYBE you could teach yourself to be an AS/400 expert by wading through said manuals...but everyone here who learned about PC administration by wading through a Microsoft manual - or, for that matter, an Intel manual - raise your hand (and we're talking actually black-and-white MANUALS here, folks, not online tutorials or knowledge base articles)...well...still waiting...Thought so.
My manager has told me that if you want to learn ANYTHING about AS/400, then you should forget about a career in anything else, because you will have to become an AS/400 expert. Who do you think told HIM that? Why, the current AS/400 experts who want to keep their salaries up and the their jobs secure! (natch)
So here's to the Cult of the AS/400. May you all fade away into well deserved obscurity.
There's easily more COBOL and FORTRAN code out there than there is Perl or Java.
Aren't barely procedural languages supposed to have more lines per code for the same task than a modern object-oriented language?:)
Is this a sigs-optional kind of place? 'Cause I am totally down with that if you know what I mean.
To have a mainframe on your desk, check out http://www.conmicro.cx/hercules This is surely true. I started my programming in 1968 on an IBM 1620 My last well-paying job was as a mainframe BAL and COBOL programmer. The company was "merged" with one in Southern California; Much of he IT in Portland went away, just a few contractors around to keep the wheels moving. I had developed an intranet warehouse application and a big-iron-monitor/program responsible/call-list in PHP on an old 386 box. Total cost to the company, $1.50 for the Linux CD (I did most of the coding outside of normal work hours). Them was the days! I went to a company to re-write a DOS application into Linux, but that didn't work out. My skill set includes FORTRAN, COBOL mainframe Assembler, JCL, MVS and DOS/VSE style systems, X86 assembler, C, PHP, Ruby, Web servers,Linux, *BSD, Networks -- you name it ... I've kept up with the technology. The fatal mistake was not jumping into bed with Microsoft.
At 60, I'VE BEEN UNEMPLOYED FOR A LONG TIME. Here in Portland you're lucky to even get a reply to a job inquiry because of the sheer numbers of applications for a job posting...
I'd gladly take a mind-numbing job as an operator (did that on a S/360 Mod 30)!
Official Pi Ambassador -- inquire for details!
I really just want to take some pictures of the thing, before they destroy it (hardware that we don't have a specific procedure in place for "data cleansing" is destroyed, and I know those guys arent going to remove just the huge disk drives and let it go at that). But I don't think the big boss will like the idea of me wandering around with a camera, so this thing will be lost completely. Damn shame. Should be sitting in a museum. Preferably *my* museum.
Well, my roomate up here in Portland is currently employed in the process of moving the county Sherrif's department off of their old mainframes and onto a more modern network. Since most of the existing personnel are vehement 9-to-5ers who have no interest whatsoever in learning how to manage the new system, there is a good chance they will find themselves looking for work at the end of the project.
--The Cyberwolfe
Ahh, I see you've decided to go psycho. Godspeed.
My personal experience with mainframes literally goes back to 1981. I have been working at the same company since then.
I worked on an NCR Criterion which had the unique ability of being able to load a boot floppy and come up as an NCR or an IBM machine. We had two of them, one was the online system (connected to cash registers that served as terminals, and thermal printers) and one that ran batch with 4 data entry terminals. Data transfer between the two systems literally consisted of lifting removable hard disks and remounting them. They each had something like 1-4 MHz and an amazing 1024K, 315 or so of which was used for the OS and not available for jobs.
Operators literally had control of the machine, and regularly had to do hexdumps on screen to help programmers troubleshoot.
For all of that we were able to run daily batch, payroll for 1500 people, and also run materials management AND service bureau-type work for a supply company.
It was a cowboy environment with all sorts of oddball characters and shenanigans I do not care to put on the web. A perfect opportunity for a rebellious type like me to slide in and escape minimum wage.
We moved to a Burroughs machine, which had a stunning, amazing 6 MB (whoa!). It was also a multiprogramming OS (MCP) with a console that was an operator's dream.
I have worked with AS/400s, 390s, RISC, NTs, various Unixen, etc., and nothing, nothing, NOTHING compares to the power, flexibility, this tech ownz joo of the Burroughs MCP ODT. Plain english syntax that generates utility jobs out of thin air, CANDE as the purest most righteous development environment, constantly refreshing screens that were configurable and gave you useable info- magnificent!!!!!
Burroughs created virtual memory and did multiprogramming years before IBM knew how to spell it.
There were problems with Burroughs, of course. They never got the I/O optimized like IBM did, they hired the most pitiful salesmen, they didn't get the machines into the universities, and of course they merged with Sperry, creating Unisys and trashing the last major architecture to threaten IBM dominance. Once again, the superior technical company does not win.
During this period PCs moved into the neighborhood. Originally we checked into B20s, but that was a dead end and we were off and running into the PC world. As a result my job duties turned into part classic mainframe operator and part Help Desk. So I ended up troubleshooting everything on earth, from mainframes to minis to user issues.
We then moved to IBM due to a vendor choice, then we merged with a similar company in our region that also had an IBM mainframe (but home-grown apps). I knew I was taking a step back technically, as IBM takes backwards compatibility to the extreme.
In many respects IBM mainframes are like sharks, they have ancient elements built into them (I like pointing out that the 80-line limit is because those batch and data elements are literally Hollerith cards to the system), but since they do the job and more importantly do not change unless there is a culture of software stability that simply does not exist in other systems. Also, the Beowulf clusters that everyone loves are a long-time feature set of IBM mainframes (parallel sysplex, WLM, etc.).
During that time I moved out of the computer room and into the sysprog seat. I do not earn fantastic bucks and I have probably 3-4 years to go before I am a star, but I can look forward to at least a good 10 years of stability before other factors may knock my mainframe path out from under me. I'm 41, so this demographic favors me mightily.
I sometimes wondered if I missed something by not being a webbie, but it looks like the world is coming to my door. My dinosaur may have to play with penguins to grow, but that's fine by me.
________________________________________ History Must Not Fall Into The Wrong Hands ___________________________________
Schroder, being a "good German" likes to be the recipient of a Dirty Sanchez.
In college I worked part time as an assistant operator on an MVS system that took up 3/4ths of the 20th floor of a 24 story building. I didn't do anyting hard core, just grunt work - I took care of the tape library, escorted technicians, cleaned equipment.
Due to lay-offs (they called it right-sizing), they were understaffed. The management had the entire operations staff set up on 12 hour shifts with a monthly rotation. Once every couple of months it would be necessary to shuffle the shifts a bit to cover vacations etc. Which invariably meant that someone ended up working a double-back 24 hour shift. The week before I started, a senior operator died in a car accident when he fell asleep at the wheel after one of these 24 hour shifts. I've always figured they hired me to take his place.
I quit the job after 6 months when one of the managers let it slip that they were going to right-size a few more operators and wanted to know if I had any friends that could come in part time.
Sounds like hell, but these operators loved it. With only one exception, they claimed that it was one of the best companies they ahd ever worked for. The one exception was the the most senior operator, a crusty old bastard named Jack, who remembered the good old days and wouldn't have been happy in his job under any circumstances (he was right-sized a few weeks asfter I started - bad for moralle I guess).
To me, it makes perfect sence to me why people wouldn't want to be operators.
This is surely true. I started my programming in 1968 on an IBM 1620.
My last well-paying job was as a mainframe BAL and COBOL programmer. The company "merged" with one in Southern California; Much of he IT in Portland went away, just a few contractors around to keep the wheels moving.
I had developed an intranet warehouse application and a big-iron-monitor/program responsible/call-list in PHP on an old 386 box. Total cost to the company, $1.50 for the Linux CD (I did most of the coding outside of normal work hours). Them was the days!
I went to a company to re-write a DOS application into Linux, but that didn't work out.
My skill set includes FORTRAN, COBOL, IMS, mainframe Assembler, JCL, MVS and DOS/VSE style systems, X86 assembler, C, PHP, Ruby, Apache web server, Linux, *BSD, Networks -- you name it ...
I've kept up
with the technology. The fatal mistake was not jumping into bed with
Microsoft.
At 60, I'VE BEEN UNEMPLOYED FOR A LONG TIME. Here in Portland you're lucky to even get a reply to a job inquiry because of the sheer numbers of applications for a job posting...
I'd gladly take a mind-numbing job as an operator (did that on a S/360 Mod 30)!
Official Pi Ambassador -- inquire for details!
If anyone's really interested in learning about modern mainframes, I run a general interest site about mainframes. You're very welcome to visit: http://www.sysprog.net .
.
And for anyone who complains that all mainframers are boring old greybeards (not me), try http://www.sysprog.net/quotes.html
To clear up some serious confusion, mainframes are the very large business systems IBM , Hitachi and Fujitsu make. IBM calls its mainframes the zSeries (z800 for Linux and z900 for z/OS). The AS400 is a midsize computer, not a mainframe. A mainframe is to a PC as a race car is to a bicycle. Anyone still need to ask why some of us love working on them?
Please don't be put off by people who think that mainframe programmers all work in COBOL on green screens in CAPITALS. That hasn't been true for decades. I program in C and Java and use XML and so on. The difference is that mainframes are much more complicated systems than workstations. Call it a challenge, if you like.
Celia Redmore
For some reason, I've only just noticed that all the previous posters have been harping on about AS400 when the subject of mainframes comes up. Has nobody on Slashdot ever worked with any of the others? Years ago, I was considered employable as a contractor because I had experience on Sperry/UNIVAC, CDC, Honeywell, Cray, Burroughs, and a few other families of mainframes. It was considered an advantage over having an all-IBM background. What happened? I know many of the above have merged/gone under, and/or are producing *nix systems, but still...
- feed the punch cards in one end
- mount some tapes
- change the paper in the bigass 1401 band printer
- collect the cards from the other end
- it's Miller time.
In all honesty it wasn't as bad as it sounds because back then the single most important person in the Data Center were the Operators. If you tweeked us off, Hell would freeze over before we'd submit your job. If you were extra nice you might even get your stuff run with the same priority as the normal production jobs.You know, I used to be able to read a punch card just by looking at the holes. It's was a lob time ago but I still remember the fun times.
--
If I actually could spell I'd have spelled it right in the first place.
It is not surprising that Marist is trying to recruit more help, they could probably use it. So could anyone else who has a mainframe these days. Marist was actually one of the first mainframes to ever run linux. They have extremely close ties to IBM, though it seems recently its become much closer. Hence the need to recruit. As IBM deploys more and more linux-frames they are probably running short on people who can manage it; young people that is.
I was actually one of the lucky college students chosen to become a mainframe operator at Marist. I must say the system never went down during my time there. It is quite a system to see running though I would not want to do it for the rest of my life. The linux console as well as countlesss other systems just sat there and spit out audit trails as we ran simple check programs to make sure the system was up and running correclty. I think that one of the biggest reasons is the complexity. Coming from a Windows/Linux background it was extremely difficult understanding what was going on and frightening to see old school terminals. Oh yes I played with MVS,OS3,JES and countless other 3 letter acronyms and to be honest I cannot remember what any of it was for and that was 2 years ago. As a student looking to the future you immediately think of the past and no future when you see the terminals staring at you. Then you look upstairs at the type of people working there, friendly as they were, and thank god that your making more money then the rest of the financial aid workers and goto the bar every night.
As a student operator you were give a large manual as well as checklists on what checks must be run on systems during the day. There were basically custom manuals for everything and they were all out of date. The biggest running joke was the fact that they were trying to have the student operators update the manual, but no one knew what any of the commands they typed did. I would say I was one of the most competent operator's there and I never once thought of getting sucked into working as a programmer for the schoole, or continuing as an operator regardless of the pay. Not that it was any good anyway. In fact, no one did. Not a single person had any interest what so ever in working with there after college.
In short the reason that they don't have enough people is that even the people who do get exposure to them don't want to run them. Any one of us could have been hired immediately after college to work on any part of the system but everyone left and went home to look for other jobs, probably at the local diner.
How do I become a "qualified" mainframe guy?
Is a degree in computing science required? Will actual work expeirence substitute for this? Where do I go? Where do I learn?
I mean, I know many will say "you can't learn from courses".. but... as a professional systems guy.. where do I go when I want to step up to mainframes? It's not the kind of thing you can just bang together in your spare time.
First, read this so you know my background.
I moved into computer operations management primarily to maintain control of my environment and earn a measly $1.00 more an hour to start. I had been under supervisors who made bad technical decisions in my judgement, and did not like the experience.
The job was hell on earth, and largely due to the nature of people who choose to do that job. This poster got it right, it's basically a haven for people who have the intelligence to do the job and the desire to hide out from a 'normal' job. We are not talking your team players here.
The 24x7 shifts mean job security, yeah, but also the constant wear and tear on nights and weekends meant anti-social behavior is reinforced amongst people who are self-selecting for it anyway.
The fact that I was one of them did not help as I wanted to be as lazy and non-team oriented as the rest of them but could not due to my position. I did not start out as a good leader type to begin with, and had to painfully learn the craft of training and stick and carrot with many ugly lessons learned the hard way.
One of the biggest problems we had was that we could not seriously threaten termination for anything but the most grevious of errors due to the lack of suitable replacements. The systems we run HAVE to run successfully 24x7, no exceptions period. So you cannot just plug in any dweeb with six months of VB/networking at the local community college. So training means standing over them to make sure the processes get done without failure, and takes overtime and care to make sure the mission critical stuff isn't destroyed.
Getting rational reasonable operators who were good and not insane was a difficult thing to accomplish. I literally had situations where bowling alley managers interviewed for me, and later I had to ask myself if I wouldn't have been better off hiring them instead of the jerk we got.
I am even now having to deal with the operator conundrum as a sysprog as some new guy screwed up our monthly database reorg apparently because he thinks he is a genius and understood his instructions without asking or calling.
The solution for our company re: replacement has been to outsource for new operators, try them out to see if they actually know what they are doing, and hire them if they work out.
There IS an operator advancement process at our company- select operators have made it into my systems area and others become Operations Analysts, doing similar work but more on an operationalized basis rather then systems. The Ops analysts are sharp sharp people and just as good as many of the sysprogs. So those posters who are concerned about ops being dead end should make sure there is a similar path before hiring on.
The whole experience was probably good for me as a human being as I am more likely to be sympathetic and understanding of both sides of the management and employee experience. But I am very very soured on ever BEING in ops management ever again and I would have to be very very hungry to ever consider it.
Most ex-ops people I work with feel the same way- it's kind of like helpdesking, it's a job and someone has to do it, but we aren't planning on doing it. And that is your opportunity to grab a job.
________________________________________ History Must Not Fall Into The Wrong Hands ___________________________________
We're looking for mainframe and AS/400 people, especially with C/C++ experience. Damn hard to find.
You're on something here. Yes, that's a good analogy. Only, the technology is bleeding edge. I mean, have you seen the insanely huge processor plate in a zSeries mainframe? One 5x5 inch ceramic plate, 20 processors, goobles of cache, a dozen dedicated I/O processors, gigabytes per second of bandwidth, and the instruction flow in each processor is executed in parallel by two CPUs, and the instruction is redone in case the output disagree (which happens only if a solar flare or stray alpha particle is hitting your CPU). Try that on your spanking new Linux cluster and see how well it works during a solar storm in a high altitude location such as Denver.
Real mainframers don't light their cheap cigarettes themselves. They just turn off the air cooling unit until a few processors glow a cherry red.
--
Mad science! Robots! Underwear! Cute girls! Full comic online! http://www.girlgeniusonline.com/
i started reading the comments from and just gave up. so many of the posters are just... clueless... but how could they not be, they so obviously have never worked in an mainframe-centric data center... understandable, they're rare -- and rarified -- unglamorous environments.
i don't dispute the value of formal education, but you won't learn data center operations at college (many of which couldn't even afford an annual license for ibm z/os). i'm a veteran mainframe programmer (ibm s/370 mainframes and their descendents), i've been doing this for nearly 19 years. i don't consider the operations side of things boring, but that's just my personal view. if you're interested in learning it, my suggestion would be to hire on with a company with a large data center and then start learning about the company's functions with an eye toward moving yourself into the ops center.
That Ace's Hardware link everybody connects to is a good overview of the IBM architecture and components, but frankly there are plenty of other machines that to my mind are mainframes as well.
Here is a definition that is as good as any, just keep in mind that there are plenty of other mainframes that do not have the IBM label on it.
Since everyone around here loves TCO arguments, here is the mainframe bid for cost-effective computing.
A general mainframe nerd site, with great links and how-tos.
Another good link site.
The dino web ring, a master compendium of 390-related sites.
Official IBMese for mainframes, with more sales power then you can shake an MS manager at.
The granddaddy of all user groups, SHARE kicks butt, defines system requirements to IBM, opposes UCITA, and changes your world more then any 5 computer gatherings you can mention. The members of this organization RUN your bank, credit card, hospital, government and corporate systems. Join the club.
And finally, you can run a mainframe on Linux or a Mac. Warning, IBM has very strict rules about their OS licensing, you are going into uncharted territory if you do ANYTHING remotely work-related with this. But you CAN run a mainframe emulator.
________________________________________ History Must Not Fall Into The Wrong Hands ___________________________________
If you need the book for COBOL, you should seriously reconsider your career choice. It may be verbose and frustrating, but it's also about as easy to read and use as you can possibly get.
I do not fail; I succeed at finding out what does not work.
To pick up where I left off there is 45-85 million men in north america with the same button to push, some one just has to find it. Plus a whole lot of money to be made when someone finds out the howto.
This bridge is burnt...
Here's my situation. I was laid off in November. I have 15 years experience in IBM VM system administration and VM systems programming. I've installed, configured, and customized VM/SP, VM/HPO, VM/XA, and VM/ESA systems. I can write assembler. I'm basically a VM kernel (nucleus) hacker. I'm not a "grey hair" -- I'm in my mid-30s.
I've been scouring the job boards for months, and I haven't found a single freaking job posting for a VM systems programmer or systems administrator. You would think that with the current IBM push towards Linux under VM, there would be a job market for people like myself, but mainframe programming and admin jobs almost never appear in the usual tech job web sites. The few jobs that do turn up are invariably for MVS work.
Where should I be looking? Does anyone have any idea where experienced VM mainframe folk go to find mainframe jobs?
That's probably due to the microcode. There's a layer between the hardware and the OS which made that posssible.
FWIW, I'm in a COBOL-based AS/400 shop. Let me run through the ages of our core "Mission Critical" team:
25, 28, 29, 38, 38, 39, 40, 45, 57
All in all a pretty young group that writes peer-reviewed structured code, performs iterative -and- waterfall development (chosen based on reality, not ideology).
In my four years here, I've never seen any of the AS/400s crash or go down unscheduled. We run a distributed ERP environment that replicates across several facilities all across the globe and even uses Domino for several key web front ends (for when it's external user-facing, or where installing a 5250 client doesn't add as much value.
Is there bad code out there? You bet, and we drive a stake in it when we find it. The problem? Amateurish "developers" from the days when the company didn't want to pay for "real" people with real backgrounds and understanding of architecture, quality-from-the-start, and user-focused design.
Overall, it's a pretty cool environment with sharp, youthful people that work out over lunch and are truly fun to be around.
Compare that to our parent company, which is a stuffy Microsoft shop, whose weekly service reports exhaustively detail the myriads of BSODs, hardware failures, mystery performance issues, and unscheduled server bounces they have to perform each week.
And they think we're backwards due to a piece of hardware. Tisk, tisk. Amateurs.
I eventually got dragged into mainframe work which was a new low for me. Believe it or not, my company has network operations and mainframe operations grouped together. I was more into learning the network operations side of things, but unfortunately I had to learn all the mainframe stuff to. I'm not extremely intelligent with the mainframe side of things, but my company didn't set it up as a learning environment.
I can see mainframes becoming a thing of the past, because all the jcl programmers (vse mainframe, I think they were IBM AS/300 mainframes), never taught us anything, we were taught to be trained monkies and just know how and when to run certain jobs. As much "extracurricular" work I put into learning about VSE I was still unable to learn fully about the JCLs that were company specific. I think all the current mainframe programmers are pretty secure in their job holdings, but at least in my company, they weren't continuing to educate us younger crowd on how to program, etc.
Bottomline is, mainframe operator work alone is probably one of the most god-awful jobs, they don't have to pay you much, because you don't necessarily have to be computer-literate, you just need to learn commands and keystrokes, the work is tedious, because you have to monitor CPU usage, and monitor jobs that can run along one another.
But as I said,, that is my own personal experience and my companies situation, YMMV.
---
Mike
I'm going to kick the next person that I see with their karma rating in their sig.
Sounds like this employer's trying to discriminate against older computer professionals (a BIG problem if you're over 40 and just want to program computers for a living). Saying that you can't find young ones just means he can't find cheap ones.
Sorry, those sub-wage H1-B Chinese and Indian kids aren't qualified. Maybe you'll have to actually pay a decent wage for your computer operator!
Best Buy can have you arrested
Where to start! Since my first job Ive had to learn 28 programming languages, 14 operating
systems, Ive dealt with paper tape, metal-oxide
tape, punched cards of all sizes, crappy 3270
terminals, modems starting at 300baud, avast array
of printers, I helped install 4Meg of memory on
a 370/148 CPU that was the size of 4 refrigerators, Ive worked on JES and HASP, read
dumps for a living, spent a few years applying
maintenance to MVS ( old SMP and new SMP), worked
on FEPs that supported a dozen 64k lines ( fast!!), seen disk storage go from 300Meg to terabytes ( peta on the way!),worked 36 hour shifts, worked from 5am to 6pm for 4 months ( 6 days a week), wrote too any assembler programs
( still dream about 0C4s). 6 years ago I switch from old iron to network engineer. Now Im adding
cisco junk to an ever-increasing list of 3digit
acronyms.
You know what? Im pooped! This computer field is
a young mans game. You guys got it. Im outta
here. Opps, cant retire yet. Im broke....rats!
Best description of mainframe environment I have seen for a long time. The original question of operators - skilled but not very well paid. Best way to learn whole systems not just some programming. There you will see the whole pain when application people only see their small pieces. You will never forget that and you start understanding why it is important to look the whole picture (the whole corporate as a customer). After trying to figure how to get those one million bills printed for next morning when someone has designed the system to change the printer font, whatever ( earlier the type of paper ) after every bill - no way in 20 days ! That's where the system programmers come - 1am working with operators they fix the application and everybody is happy in morning. Systems programmers and operators have actually much more common than systems programmers and appication people. And I agree with your mail - any Windows / Unix installation seems unorganized and very inefficient after working in real mainframe environment. And AS/400 is not mainframe even very nice system - mainframes are systems that serve tens of thousands of users 7x24 while happily processing huge transactions millions a day - not just those small banking transactions. And fast.
I work for a huge datacenter as a mainframe computer operator. Yeah, the guys doing the systems work are mostly old - 50 years plus, however we have a couple guys on that side who are 30's, 40's - previous operators who got promoted.
but the vast majority of the operations staff is 20-somethings, right out of college or still attending college. We have ONE guy who is 50 something in operations, and he has been operating for many many years, yet still is one of the worst unskilled guys we have. I've been doing this for 3 years and I could run circles around the guy when it comes to troubleshooting and operating MVS or VM. Generally, operators spend a couple of years in operations, then get promoted out, usually these days to the unix side as junior sysadmins. But maybe I will break the mold and be the one 20 something to get promoted and actually stay with the mainframe
this isnt like most places, where you have one system to run, our customers are state government agencies who due to budget cuts had to outsource their mainframe and high end unix IT. We take in their systems and run them, so we have such a variety of systems, it is really a pleasure to work and learn with.
most of our clients are S/390 or z/ running OS/390 2.10, we also have a couple of VM and VM/ESA clients. Some agencies went the Unix route, so we have some Sun enterprise 10k, IBM AIX 64bit massively SMP boxes, etc. But our bread and butter is still the big iron.
It has been more than 15 years since I was a mainframe operator (IBM 360/50 running EDOS/VS) at the university. We did all the business processing for the school (accounting, payroll, alumni stuff, student and course tracking, etc.) and some local payroll and mailing lists as well.
I was trained apprentice style and it took more than a year because you don't SEE some jobs more than once in a year and some jobs had their own quirks. Plus some programmers (or rather JCL coders) would mistype their JCL and you had know if you should call them if the job (not the OS) bombed so you could change it and rerun it or just stuff it in their bin. The bigest fear was a power outage (we couldn't afford a backup generator) from electrical storms or kamikaze squirrels who take out transformers (it really happened). When that happens you are suddenly sitting in the very quiet darkness.
It could be mind numbing but as a student it was a blessing. I ran my class project programs (Assembler, FORTRAN, COBOL), anytime I wanted (everyone knew me in the center) while most students had only two runs a day. In addition, when we got a DEC PDP11/45 there was a terminal in the machine room as well (I had to do backups) so I had unlimited access there as well.
The mind altering part is the work consisted mainly in loading decks of punch cards, changing disks (20mb in huge 15 or 20 inch multi-platter packs), tape (9 track 1200 bpi), and paper; and it was mostly paper. Some jobs cranked for hours and spit out two pages (header and EOJ). Others ran quickly and printed all weekend (end of year general ledger). But printing 5 boxes of paper (with no, or 1-5 parts i.e. up to 4 carbons) a weekend was about normal. But that still leaves lots of hours sitting in the room LISTENING to all those blowers in a room illuminated by floresent lights watching the LEDs flicker on the panel. It got to the point that I could tell when a job was finishing up by the pattern flashing. In fact there were many times I dozed off only to awake JUST before a job finished up. No ESP, just the sound of 9 track tape drives rewinding....
Perhaps these people should not be searching in the IS field because you don't need to know the theory and you can teach this to anyone who wants to keep a job.
The old ones I used back in the mid 80's were huge. I think it was purchased in the 70's.
Five years ago I was part of a very large project replacing a substantial COBOL development on mainframes with an n-tier C++ system. There were about 15 COBOL staff and plenty of ribbing between us and them. We ended up getting one of those huge Jurassic Park banners from the local cinema and hanging it over their work area :-) Of course management made us take it down...
Never, ever lose a file again. Ever.
The mainframe environment varies widely from company to company. And then people who have only seen a bad mainframe political environment (where people have not even absorbed newfangled ideas like "functions") complain about their environment and think that they are talking about a universal phenomena. Which they aren't.
However they are talking about things which won't get fixed in their current environment any time soon. And even if you do get a willingness to change, you have decades of code guaranteeing that headaches will continue indefinitely.
So the problem is real. Just the cause is misidentified as being mainframes in general when it is more specific to their company politics.
*tap* *tap* *tap* *tap* *tap* *tap* Jesus Saves! *tap* *tap* *tap* *tap* *tap* *tap* *tap* *tap* Jesus Saves! *tap* *tap* *tap* *tap* *tap* *tap* *tap* Jesus Saves!
The Son of God using Microsoft Word
I have just landed a great job using Fortran - forget COBOL, Fortran is the way to go.
I stole this
I work for a company running a good ol' IBM Mainframe and I programmed on it for 2 years. I honestly wanted to hurt myself everyday because the job was so boring. Fixing Y2K programs in COBOL that have been written well before I was born is sick and wrong.
Also, I think the main reason doing mainframe work is so hard is that you can't really see your results. If you do PC programming or Web development, you have a nice interface that you can show off and say, "Look, I made this!", but with the mainframe, all you have is an ugly Green Bar report or a text file. Where's fun and excitement in that?
All I have to say is that two years ago when I was looking for a job I applied to about 4 operator positions. I was rejected from ALL of them only because the company did not want to train me. They want people that they can pay peanuts and don't have to train. They do not realize that this isn't going to happen. Thankfully I have moved on to a better job as a Unix Analyst / DBA / programmer (for simple projects). If they don't want to spend the time and money for training, of course there will be a shortage.
Sure, there is a lot of people seeking people with AS/400 and mainframe knowledge. Stuff like Baan and SAP is also something it's good to know to find a nice job with a confortable salary. But... How to learn AS/400 administration? How to learn SAP? I don't know of any general computer science school that teaches those. Learning centers can deliver these formations, but they are really too expensive for an individual. I had a look at IBM's curses, and they are also way too expensive. And self-formation is impossible without the hardware/software that is also out of price. Some months ago, I was proposed a very interesting job, but some knowledge of AS/400 systems was needed. I tried to convince them to pay my initial formation, but they refused, saying that a support contract with IBM would finally be a better and cheaper alternative. So there's a paradox here. Recruiters need people with this specific knowledge, but there's no way for people to get the needed skills.
{{.sig}}
Lets see this type of migration with Microsoft. You have to rewrite your applications every four years.
I work on an AS400. Its definately not all green screen. The main problem as I see it with AS/400 is that because its actually 'possible' to be running code thats 20 years old on a Power PC Processor with 64bit objects that people DO! Unlike Windows based apps there is no requirement to rewrite the code. And as many people have mentioned this is a business machine, and no business case exists to rewrite code so it can be 'prettier'.. Really.. An intuitive interface can be anything. I've seen my fair share of shit GUI interfaces - the same as I have seen shit GreenScreen interfaces. Its all about the programmer.
:)
I use eclipse ide to develop in RPG. I think IBM threw a pile of money at this to help get it started, but its hot dev tool. Looks like you could do just about anything inside it.
And for supa mod points - can I just mention that its possible to run Linux on the AS/400 in a logical partition! =-- See - mentioned Linux - Mod me UP!
Sorry for rant. I know the topic is finding operations staff for this machine. So I offer myself! If anyone requires one really bad and wants to pay me lots of money.. well..
Brian
I heard people grumbling about this and the general feeling is that IBM is encouraging this shortage.
They theory is that when a company can't find enough people they will outsource it. A big outsourcer is IBM Global Services, so they can make more money. IBM can more easily train people since they are so big.
Just a rumor, but an interesting one...
3 words? Mind numb ing? I can easily see your problem. Pity you can't.
Laugh at mainframes and COBOL all you'd like. Someday, if you gain any wisdom, you may figure out how foolish those comments were.
security: have you ever heard of a security break in a mainframe? any script-kiddie trying to break in ? any RACF/SNA cracker around? development tools: lots (eg. FileAID, ISPF) languages: Lack of languajes? cobol, PL/I, perl, c/c++, java, rexx, jcl long-life: a dyin technology? bullshit. Its being used widely nowadays (insurance, banking, public services, gov. agencies, etc.) I've worked on a ES/9000 (MVS/TSO, some VM/CMS) for about a year, doing some JCL and cobol stuff. Its an amazing technology. nothing can be compared to the good'ol mainframe.
- This can't be... - Be what? Be real?
I'm 23, got a BS in CompSci, minor in math and psychology, and I work on a 400. After graduating back in May of 2002 with a good GPA, ACM leadership, and an internship and coop. It still took me 6 months to get a interview. Most of my background is C++ ,Java, and Perl, lord knows that I would have ended up hacking RPG after school. The company I'm working for needs to provide 99.9% uptime to electrical, water and gas companies and the 400 does it like a dream. It took me about 2 months to pick up what was going on but now I only have to ask questions that more design issues than logic or semantics. It's not a dream job, but heh, a job right now is fine with me. I am our youngest programmer by 17 years by the way. Suprisingly the majority of our staff is women too. ( Who have daughters that drop by to say hi frequently)
May I please have my frontal lobotomy if I bring back the ashtrays?
We had a variety of IBM mainframes at SUNY Binghamton (NY) as a result of grants from the local IBM facilities. They were much more powerful than the various VAX machines spread around the departments. After school I went to work for IBM on various projects that ran on some seriously huge RS/6000 machines (not mainframe, obviously) as well as the big iron. VM/CMS and MVS were the systems of choice for business at that time and I'm very sad to see them go away. The colorful text-screen oriented IBM 3270 terminals were absolutely perfect for data-entry and point-of-sale terminals. Much later I found Windows machines at other companies that just ran IBM 3270 emulators to connect to these same mainframes that were running 'way back then.
:)
The move to client-server has been very expensive--I don't have quotable sources but you can search on studies done on the cost to businesses of using client-server methodology vs. terminal-mainframe and in many cases the mainframe was cheaper, more reliable, easier to run, the list went on. I believe SciAm did a piece on this issue back in the mid-90's when client-server was in its hey-day.
And, heh, in case anyone is facing the attrition demon, I would be happy to transition back to the mainframe world and put my VM/CMS, MVS, JCL skills back into use at your firm
Kris
Kriston
Soon there will be reports about a lack of skilled netware pros. Better shine those CNE pins!
I've determined that Moore's law is not driven by technical innovation, but simply by the need to keep up with shitty programming.
:)
I think closer to the mark is the need to keep up with shitty *requirements*.
Or so says this shitty programmer.
You *do* know that there's a Gnu COBOL project?
And do let me know when I can have a single box with 100 PCI-X buses and memory architecture which can support the load without seriously starving the CPUs. Mainframes are a whole 'nother world.
Thank goodness Unix has something vaguely resembling real mainframe-style batch jobs. I'm forever trying to cobble together something on our Windows boxes to let me say, "take care of this. Let me know when it's done. And show your work." as in the good old Bad Old Days when computers worked for us instead of us working for them. A typical workday these days reminds me of the older robots in Asimov's "Runaround", which could not move unless a human was riding them. There is definitely value in some of those old ways.
In 98 I graduated and got a job using System/36 environment on an AS/400. I was working with an accounting system that was over 20 years old. Documentation for System/36 was never around, it was hell. RPG/36 code is horrible! RRG 400 isn't that much better.
So I moved on to Java. Built incredible and beautiful applications. I was in heaven.
Then the company went bankrupt. So now I'm back in the System/36 environment running on an AS/400.
Kill me now?
I started with mainframes some 30+ years ago - then switched to PCs just before CPM came on the scene - then switched back to zOS (mainframe) last year. Much to my chagrin, not much has changed - there's just more of it. The machines are BUSY and various user programs written in (for instance) SAS run as frequently as 10,000 times each day. Transaction based systems process millions of transactions an hour. The databases (like DB2) hold hundreds of millions of records.
What surprises me the most and what I think is the biggest issue with finding new people to support both/either the OS or the application programming is the fact that the majority of interface takes place on a 24 x 80 character green screen. How many college courses teach ISPF (the green screen interface) - probably NONE.
Besides the complexity the impact of any change, improperly made, is enormous. You screw up a line of code and a million people don't get their welfare checks on time - or worse yet they are all mailed to the wrong address.
The old timers are getting ready to retire (me included). It will be interesting to see what falls out. I imagine we will see a couple of years of disaster and then a fairly sizable increase in money for mainframe expertise.
Click here or here.
I'd rather go hungry than do COBOL on a mainframe.
The problem with mainframes is that the younger generation is going to have to eventually inherit the older generations problems. Which is a fundamental problem. You have the older grayhairs who don't want to get axed before they retire, and you have the younger hotshots who couldn't give a hill of beans about a decrepit and outdated kluged monstrosity. I graduated about two years ago, and I inherited a VAX/VMS based system for a power plant. Yeah Vaxes are old and klunky, but there's something to be said about a computer system that has been running a nuclear power plant for over 12 years. We have over 8000 gizmos hooked up to a computer that has a little bit more computing horsepower than a P33(yes that is a 33MHz not 133 even). I know that this probably isn't everyone's cup of tea, but I find that everything I've learned here is going to pay off it's weight in gold in years to come(5-15 years). Most people in this industry are getting reasonably close to retirement which means that all the knowledge I've gained now will be in high demand later. And that's the key to putting up with these old monstrosities now. Hardware doesn't run forever. So eventually that old computer will get replaced, and you'll then be the one on the head of the technology curve. The trick is looking at the worth of what you're currently doing and see how it will profit you in the future.
I ran a couple of AS/400's for a while. They were extremely important systems. I think the interaction is fascinating. There's nothing like booting 500 workstataions off and watching them slowly fall off the box. It's like a human system, almost. Everything has to get done before it just quits. And all the god damn interfaces, lines, fuck it. What am I saying? AS/400 is a bitch and needs to die.
Cool! Amazing Toys.
What are mainframe operating systems written in?
C like UNIX? Assembly like the really old dinosaur computers? Visual Basic in some fit of irony?
The fact that UNIX, DOS, and Windows are all written mostly in C, C++, and a bit of assembly is common knowledge, but it seems mainframes are more than figuratively a "black box".
Healthcare article at Kuro5hin
I WANT TO BE A MAINFRAMER! I just cant get anyone who has one to teach me. It's not like $50K boxes are easy to come by, second hand, at affordable prices. Why not take a noob and apprentice him in the dark, sunlight-deprived ways of the ubersys? WHAT"S WRONG WITH ON THE JOB TRAINING?????
arg....
Isd that you have younger guys/women who fear or refuse learning the older systems, and then you have older guys/women who fear or refuse to learn the newesr systems. Then you end up with a void in the workforce where the elder programmers and admins leave and noone knows how to run the older systems, or else newer technology is needed and the older people do not know how to even begin to implement it into their current systems. This is why we still have inefficient IT departments at nearly every company. The fact is simple. Companies need to make sure that their staff memebers get constant training to keep their skills current.
-Cnik
Some of the most interesting work I've done was inside a set of 30-year-old transaction-based applications running on a Unisys 2200 mainframe system (it was the primary flight planning and real-time flight tracking system at a major US airline).
No way they're going to replace that system any time soon. They might migrate subsystems to other platforms, slowly, but so many of the airline's business rules are so tightly interwoven with (and key operational elements are so highly dependent on) that code that porting it is going to be a real interesting process...
Mainframe/UNIX Bit Twiddler and long time Windows/Linux Hobbyist.
The Theorem Theorem: If If, Then Then.
A useful response. Someone mod parent up.
I am 23 and very interested in mainframe technology. Problem is that last mainframe accessible for students in my country was shut down two years ago. Now there are only 2 mainframes at LDZ (Latvian Railway) but they don't have personnel shortage. In fact, nobody has personnel shortage in Latvia :(
The only "mainframe" I have access to is Hercules on my Athlon XP. Hercules runs MVS 3.8j and Linux S/390 ok, but it's impossible to get any documentation or manuals for MVS, OS/390 etc.
I'd like to get any help, advice or manuals from slashdot people with mainframe experience.
Oldtimer: Ok, put the keyboard down and back a way from the VAX!
Kid: What?! All I was going to do was click this red "break" key. What does that do anyway?
Several I talked to said straight out they only wanted recent experience, a gap of a few years....meant I couldn't be trusted with the machines I've worked around for 20 years.
They wanted *recent* mainframe experience? What a hoot. Stupid PHB's.
"Wanted: Studabaker auto mechanic. Must have recent experience."
Table-ized A.I.
Indeed. There is not a shortage of mainframers, just a shortage of cheap young mainframers, and this is a "corporate crisis" that congress must of course "fix".
Table-ized A.I.
Those were the days...
I implemented a search algorithm that searched 300000+ registers in ALGOL. To speed up things I indexed the whole thing and to save disk space I compressed using an algorithm I, grasp, learned in the university. Databases? You hippies, mens in those days did not use databases (too expensive, too novel).
It is a pity that the computer stopped working after a mouse found its way in the "dish washers" (big cilynders that used to contain the hard disk plates).
Let me tell you, it was a messy affair.
IANAL but write like a drunk one.
I worked for IBM in one of their 390 labs right after I graduated from college. All I can say is that I can perfectly understand that nobody with a good education in computers would want to work on these machines.
...
/370 /390.
The OS is pre UNIX, i.e. no hierarchical filesystem and files can not change their size dynamically. When you want to edit a new text file you have to create a file of defined size. If you run out of space you have to create a new file, copy your contents over and delete the old one. But this is not exactly true someone might say! Admitted. You can actually resize your file 3 times by an increment that must be specified at file creation. Very intuitive.
Oh and the same people complaining about the filersizeing being not correct might say, that there is a hierarchy. Well, somewhat. It is possible to create a file that acts as a container for text-file only. So if you want to write a program and use several source files this is the way to go.
There also is no shell. Whenever you want to run a program you have to write a script in a completely cryptic job description language. There you have to specify how long the job should maximally take,
So after spending 15min on writing your job-script you can finally run an ls command.
Now all this isn't quite true either. What I described is only true for the basic OS. You can purchase several layers of stuff that uses DB2 databases to create something like a filestructure and so on that makes things a little easier.
The next thing is that the 390 don't have floating point units which makes them completely useless for any scientific application or even simple numerical tasks.
Oh, did I mention that non of the 15 layers actually implements a GUI?
Oh and the speed of these machines is ridiculous. Even if you have 1 processor completely for you alone it's about half as fast as a comparably clocked PC.
One of the funniest things that happened to me was when I confronted my manager with the fact that OS/390 is not an API based OS (there is no defined set of C (or COBOL/FORTRAN functions for that matter) that can be uses to access OS features from applications. He managed to find a marketing brochure from the 70 that stated that OS/370 was an API based OS just like UNIX. The brochure defined API = Assembler Programmer Interface -- isn' that funny?
Did I mention that hard disks (DASD in 390 lingo) are not block devices as we know them? No, the OS exposes information about the number of tracks and segments on certain disk. The application programmer can then write code on how to write data to the segments. If you want to optimize file IO of your app you consult all the handbooks of all harddrives in the past and figure out how to best interleave write requests...
So if you're completely backwards and want to work in a stuffy, conservative environment with incompetent managers, then the mainframe word might be for you. If your into databases and corporate computing and don't want to work with the dinosaurs find a place that uses either Suns big UNIX servers or IBM Regattas. Both nice machines and UNIX got almost everything right that IBM got wrong with OS/360
I was totally frustrated working on mainframes especially if you want to do some real programming.
One interesting factoid: If you go into a supermarket that is using IBM checkout equipment, many times you'll find an AS/400 in every 6th aisle or so. Probably under the moving belt, behind some panel. There may be a small window where you can see a green light. Quietly working away, keeping the cashiers happy.
They are configured in such a way that if a unit goes down, the workload between the aisles is re-shuffled between AS/400s (this may cause the cashier to re-scan the last few items, but usually is invisible). This reshuffling can also be done between closely-located stores. 4800 bps leased lines can handle the traffic, though you may notice a delay between a scan and the price appearing on the display.
AS/400's are not totally buttletproof. But in this configuration it's not unheard of for a machine to be down for a day or two before someone comes out to fix it (the failure is detected quickly, but if there's a few extra machines still available there's generally not a rush).
A study last year by Meta Group Inc. in Stamford, Conn., found that 55% of IT workers with mainframe experience are over 50 years old.
They say that like it's a bad thing. My mom is an unemployed 53-year-old DBA of DB2 on OS390. It seems a waste to train a younger crowd on this stuff when there are still people like my mom who need to eat!