Discussion:
Huge datasets!?!?!?!?
(too old to reply)
J. Clarke
2020-05-12 03:25:15 UTC
Permalink
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)

ARRRGHHHH.

Sorry, just needed to vent.
r***@gmail.com
2020-05-12 03:55:25 UTC
Permalink
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Presumably the files need to be on a hard drive(s) somewhere initially.
Dan Espen
2020-05-12 03:59:06 UTC
Permalink
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.

I'm guessing mainframe dasd is still relatively expensive.
--
Dan Espen
J. Clarke
2020-05-12 05:42:56 UTC
Permalink
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.

In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Dan Espen
2020-05-12 12:29:44 UTC
Permalink
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.

If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.

Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
--
Dan Espen
J. Clarke
2020-05-12 21:07:41 UTC
Permalink
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.
If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.
Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Dan Espen
2020-05-13 00:13:18 UTC
Permalink
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.
If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.
Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Most of my life, I was paid to find solutions.
No chance of that changing before the end.

Anyway, yep, throughout my career I'd run up against the computer
center creating obstacles. I had some friends there too, but
management loved rules.

Back when I started we had the old crustly EAM guys running operations.
You could make their life easier and get on their good side.
--
Dan Espen
J. Clarke
2020-05-13 01:03:12 UTC
Permalink
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.
If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.
Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Most of my life, I was paid to find solutions.
No chance of that changing before the end.
Anyway, yep, throughout my career I'd run up against the computer
center creating obstacles. I had some friends there too, but
management loved rules.
Back when I started we had the old crustly EAM guys running operations.
You could make their life easier and get on their good side.
The crustly guys were fine. They didn't understand what we did but
they understood that we have deadlines that we do not control, the
missing of which will result in Very Important People becoming Very
Unhappy. But they're mostly gone now and we have a new crop to break
in.
Peter Flass
2020-05-13 01:07:36 UTC
Permalink
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.
If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.
Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Most of my life, I was paid to find solutions.
No chance of that changing before the end.
Anyway, yep, throughout my career I'd run up against the computer
center creating obstacles. I had some friends there too, but
management loved rules.
Back when I started we had the old crustly EAM guys running operations.
You could make their life easier and get on their good side.
I had one boss who insisted we keep source programs on cards. Finally I
loaded one program to a PDS, did a lot of editing in TSO, and then punched
a new deck when I wad done. Shortly thereafter the rules changed.
--
Pete
J. Clarke
2020-05-13 01:33:12 UTC
Permalink
On Tue, 12 May 2020 18:07:36 -0700, Peter Flass
Post by Peter Flass
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
Post by Dan Espen
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Perhaps you can NFS mount a couple of TB on the mainframe and
run your application that way.
Why? Maybe your laptop has a couple of TB of RAM, but mine doesn't
and neither does any other normal person's.
In any case, if I need to use a couple of TB of DASD to get my work
done then I use it and the mainframe people just have to suck it up
and deal.
Post by Dan Espen
I'm guessing mainframe dasd is still relatively expensive.
Which is irrelevant because it's all on "tape".
Well, I guess I missed the "tape" part, all you said was huge datasets.
If they had objected because of disk space, I was suggesting you load up
a PC with a couple of TB and NFS mount that on the MF.
Tape is another matter, they want multiple jobs to handle tapes.
One job per tape?
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Most of my life, I was paid to find solutions.
No chance of that changing before the end.
Anyway, yep, throughout my career I'd run up against the computer
center creating obstacles. I had some friends there too, but
management loved rules.
Back when I started we had the old crustly EAM guys running operations.
You could make their life easier and get on their good side.
I had one boss who insisted we keep source programs on cards. Finally I
loaded one program to a PDS, did a lot of editing in TSO, and then punched
a new deck when I wad done. Shortly thereafter the rules changed.
We did some of that at UCONN when I was in grad school. There was one
idiot in IIRC the Chemical Engineering department who insisted that
his students use cards because (in the mid-80s) that is what they
would have to do in "the real world". There was one keypunch and it
was balky enough that none of us wanted to deal with it, so we taught
the students how to edit on the terminal and punch a deck for the
idiot.
Charlie Gibbs
2020-05-13 03:55:13 UTC
Permalink
Post by Peter Flass
I had one boss who insisted we keep source programs on cards. Finally I
loaded one program to a PDS, did a lot of editing in TSO, and then punched
a new deck when I wad done. Shortly thereafter the rules changed.
Small Univac shops like ours always kept source programs on cards,
although a bit of study of available utilities convinced me that
this wasn't necessary. One day, the new department head, who had
come from an IBM shop, was musing, "Why couldn't we keep source
programs on disk?" Seizing the opportunity, I joyously cried out,
"But we can!"

I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future. He did say
something equivalent, though, and all was right with the world.
--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <***@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.
Thomas Koenig
2020-05-13 05:21:46 UTC
Permalink
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
Peter Flass
2020-05-13 13:58:33 UTC
Permalink
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
I miss Picard. I just watched the first season of the show on CBS All
Access with a free trial. I’d like to see more, but I don’t want another
streaming service.

CBS had (has) a free one-month trial. Sign up, then cancel immediately to
eliminate worries about being billed. You still get the whole month. A
month was enough for a season of _Star Trek, Picard_.
--
Pete
Charlie Gibbs
2020-05-13 18:54:55 UTC
Permalink
Post by Peter Flass
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
From a collection of TNG Christmas carols:

Oh, the vacuum outside is endless,
Unforgiving, cold, and friendless.
But still we must boldly go...
Make it so, make it so, make it so.
Post by Peter Flass
I miss Picard. I just watched the first season of the show on CBS All
Access with a free trial. I’d like to see more, but I don’t want another
streaming service.
CBS had (has) a free one-month trial. Sign up, then cancel immediately
to eliminate worries about being billed. You still get the whole month.
A month was enough for a season of _Star Trek, Picard_.
We're currently watching all of TNG from the beginning on Netflix.
--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <***@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.
jtmpreno
2020-05-14 11:51:28 UTC
Permalink
Post by Peter Flass
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
I miss Picard. I just watched the first season of the show on CBS All
Access with a free trial. I’d like to see more, but I don’t want another
streaming service.
CBS had (has) a free one-month trial. Sign up, then cancel immediately to
eliminate worries about being billed. You still get the whole month. A
month was enough for a season of _Star Trek, Picard_.
CBS is pulling the plug on CBS All Access:

https://bingenews.ca/2020/05/07/cbs-pulling-plug-on-all-access-in-rebranding-effort/

I would have liked to see Picard but I already spend too much money on TV.
Peter Flass
2020-05-14 14:01:26 UTC
Permalink
Post by jtmpreno
Post by Peter Flass
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
I miss Picard. I just watched the first season of the show on CBS All
Access with a free trial. I’d like to see more, but I don’t want another
streaming service.
CBS had (has) a free one-month trial. Sign up, then cancel immediately to
eliminate worries about being billed. You still get the whole month. A
month was enough for a season of _Star Trek, Picard_.
https://bingenews.ca/2020/05/07/cbs-pulling-plug-on-all-access-in-rebranding-effort/
I would have liked to see Picard but I already spend too much money on TV.
Right, this is crazy. A service might have one show you want to watch. Is
it worth subscribing to ten streaming services for one show each? I thought
cable was expensive!
--
Pete
David LaRue
2020-05-14 17:30:29 UTC
Permalink
Post by Peter Flass
Post by jtmpreno
Post by Peter Flass
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
I miss Picard. I just watched the first season of the show on CBS
All Access with a free trial. I’d like to see more, but I don’t
want another streaming service.
CBS had (has) a free one-month trial. Sign up, then cancel
immediately to eliminate worries about being billed. You still get
the whole month. A month was enough for a season of _Star Trek,
Picard_.
https://bingenews.ca/2020/05/07/cbs-pulling-plug-on-all-access-in-rebr
anding-effort/
I would have liked to see Picard but I already spend too much money on TV.
Right, this is crazy. A service might have one show you want to watch.
Is it worth subscribing to ten streaming services for one show each? I
thought cable was expensive!
To CBS and other networks, just put in on TV where we can see it. No
need to pretend the masses will justify a custom streaming service.
Scott Lurndal
2020-05-14 18:51:51 UTC
Permalink
Post by David LaRue
Post by Peter Flass
Post by jtmpreno
Post by Peter Flass
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
I miss Picard. I just watched the first season of the show on CBS
All Access with a free trial. I’d like to see more, but I don’t
want another streaming service.
CBS had (has) a free one-month trial. Sign up, then cancel
immediately to eliminate worries about being billed. You still get
the whole month. A month was enough for a season of _Star Trek,
Picard_.
https://bingenews.ca/2020/05/07/cbs-pulling-plug-on-all-access-in-rebr
anding-effort/
I would have liked to see Picard but I already spend too much money on TV.
Right, this is crazy. A service might have one show you want to watch.
Is it worth subscribing to ten streaming services for one show each? I
thought cable was expensive!
To CBS and other networks, just put in on TV where we can see it. No
need to pretend the masses will justify a custom streaming service.
Ah, but with streaming they get 100% of the revenue, rather than
just a small piece from the cable/satellite provider, and even
less from OTA. CBS doesn't care about the viewer, they care about
their profits.

Besides, since Survivor started the reality TV trend, there hasn't
been any watchable programming on either the OTA or cable channels :-)
Jan van den Broek
2020-05-25 21:00:43 UTC
Permalink
Wed, 13 May 2020 05:21:46 -0000 (UTC)
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
He probably wasn't trying to do a build, see sig.
--
Met groeten, | make it so
Jan van den Broek | Make: don't know how to make it. Stop.
***@xs4all.nl |
----------------------------------------+---------------------------------------
Quadibloc
2020-05-25 21:50:07 UTC
Permalink
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
It's perfectly good English: "make it so" means: cause it to be true.

The meaning of "make" used is found in many common English sentences - "Make
this door close properly", "make these pants fit". The use of "so" is slightly
archaic, but used in some phrases: "This is so", "that is not so", "We need to
find out what is or isn't so".

Usually, these days, the word "so" is used more in another sense: "the rain is
so heavy", and one says "this is the case", "that is not true" rather than using
"so" in the other sense.

John Savard
Charlie Gibbs
2020-05-26 00:39:30 UTC
Permalink
Post by Quadibloc
Post by Thomas Koenig
Post by Charlie Gibbs
I wish I could say that he then quoted Captain Picard and said,
"Make it so" - but TNG was 10 years in the future.
I always wondered where that came from. It sounds like a
mistranslation from German ("Machen Sie es so", or "Mach es so"),
where "machen", while clearly related to "make", means "do" in
this context.
It's perfectly good English: "make it so" means: cause it to be true.
The meaning of "make" used is found in many common English sentences -
"Make this door close properly", "make these pants fit". The use of "so"
is slightly archaic, but used in some phrases: "This is so", "that is
not so", "We need to find out what is or isn't so".
That that is is that that is not is not is not that so

(Definitely needs the punctuation...)
Post by Quadibloc
Usually, these days, the word "so" is used more in another sense: "the
rain is so heavy", and one says "this is the case", "that is not true"
rather than using "so" in the other sense.
<valspeak>That is soooooo last week!</valspeak>

(Gimme a break, I've had a bad day.)
--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <***@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.
Charlie Gibbs
2020-05-13 03:46:15 UTC
Permalink
Post by Dan Espen
Post by J. Clarke
Oh, forget about it. I wasn't asking for help, I was commenting on
the ineptitude of the people running the mainframe.
Most of my life, I was paid to find solutions.
No chance of that changing before the end.
I love finding solutions. What drives me nuts
is when the PHBs don't really want solutions.
Post by Dan Espen
Anyway, yep, throughout my career I'd run up against the computer
center creating obstacles. I had some friends there too, but
management loved rules.
Some of my best work was done underground.
"It's easier to get forgivenss than permission."
Post by Dan Espen
Back when I started we had the old crustly EAM guys running operations.
It made it hard to migrate files from card decks onto disk.
"If you can't see the holes, how do you know it's there?"
Post by Dan Espen
You could make their life easier and get on their good side.
Yup. Find the problems they _really_ want solved (see above) -
if you can fix those you'll probably win them over.
--
/~\ Charlie Gibbs | Microsoft is a dictatorship.
\ / <***@kltpzyxm.invalid> | Apple is a cult.
X I'm really at ac.dekanfrus | Linux is anarchy.
/ \ if you read it the right way. | Pick your poison.
Quadibloc
2020-05-12 14:07:15 UTC
Permalink
Post by J. Clarke
Which is irrelevant because it's all on "tape".
Hmm. I suppose the dataset on tape could _somehow_ be distributed in a way that
makes it seem "huge".

Without specifics, I can't be _sure_ they don't have some valid reason, but I am
inclined to agree with you, that given present hardware, it should not be
possible for efficiencies to be gained by breaking up a job into smaller pieces,
where the files involved would all fit into a normal personal computer these
days.

Obviously, if the job requires random access to one of the files on the tape,
that file should be copied to disk or even into memory first, but that's not
what they're requesting.

John Savard
Peter Flass
2020-05-12 18:10:19 UTC
Permalink
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Don’t forget you don’t have all this huge system to yourself. If you do
something that impacts one or more CICS/IMS region it’s probably not a good
thing. You have all of your flippin laptop to yourself and don’t care what
else happens.
--
Pete
J. Clarke
2020-05-12 21:18:44 UTC
Permalink
On Tue, 12 May 2020 11:10:19 -0700, Peter Flass
Post by Peter Flass
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Don’t forget you don’t have all this huge system to yourself. If you do
something that impacts one or more CICS/IMS region it’s probably not a good
thing. You have all of your flippin laptop to yourself and don’t care what
else happens.
I also don't have much data. I say "huge dataset" and you stop
reading at that point and envision yottabytes of data, not the effing
800 meg that they think is so "huge".
Peter Flass
2020-05-12 23:39:36 UTC
Permalink
Post by J. Clarke
On Tue, 12 May 2020 11:10:19 -0700, Peter Flass
Post by Peter Flass
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Don’t forget you don’t have all this huge system to yourself. If you do
something that impacts one or more CICS/IMS region it’s probably not a good
thing. You have all of your flippin laptop to yourself and don’t care what
else happens.
I also don't have much data. I say "huge dataset" and you stop
reading at that point and envision yottabytes of data, not the effing
800 meg that they think is so "huge".
Since I’m not there all I can do is suggest possible alternatives to the
“they’re all dinosaurs” theory, which may well be true - I’ve seen it
myself. Half a 3390 volume doesn’t sound “huge” to me either.
--
Pete
J. Clarke
2020-05-13 01:20:36 UTC
Permalink
On Tue, 12 May 2020 16:39:36 -0700, Peter Flass
Post by Peter Flass
Post by J. Clarke
On Tue, 12 May 2020 11:10:19 -0700, Peter Flass
Post by Peter Flass
Post by J. Clarke
The mainframe guys at work want us to break up a job into multiple
pieces because of the "huge datasets". I can fit every dataset that
this job uses or produces in RAM on my crappy laptop, simultaneously.
I can even fit multiple copies of all of that in RAM on my crappy
laptop. I know bloody well that the mainframe will handle it all.
What bothers me is that the people running it seem to be stuck in the
'60s or something--they don't seem to understand how effing _huge_ the
machine they run has become (it's a current generation Z--actually 3
of them)
ARRRGHHHH.
Sorry, just needed to vent.
Don’t forget you don’t have all this huge system to yourself. If you do
something that impacts one or more CICS/IMS region it’s probably not a good
thing. You have all of your flippin laptop to yourself and don’t care what
else happens.
I also don't have much data. I say "huge dataset" and you stop
reading at that point and envision yottabytes of data, not the effing
800 meg that they think is so "huge".
Since I’m not there all I can do is suggest possible alternatives to the
“they’re all dinosaurs” theory, which may well be true - I’ve seen it
myself. Half a 3390 volume doesn’t sound “huge” to me either.
Just a new guy who hasn't learned who the players are yet. The old
guys know "It's Spring, the quants are going to bomb the system for a
week or so".
John Levine
2020-05-13 01:36:49 UTC
Permalink
Post by J. Clarke
reading at that point and envision yottabytes of data, not the effing
800 meg that they think is so "huge".
Since I’m not there all I can do is suggest possible alternatives to the
“they’re all dinosaurs” theory, which may well be true - I’ve seen it
myself. Half a 3390 volume doesn’t sound “huge” to me either.
Anyone who thinks 800 meg is huge is definitely a dinosaur.

I mean, that's about 1/5 the size of the USB keys I get as party
favors at meetings.
--
Regards,
John Levine, ***@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly
Loading...