>http://video.canadiancontent.net/5-molson-i-am-canadian.html
>From about 10 years ago.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email
>No, I've already gotten in trouble for that one!
>The "zed" world doesn't understand.
I wasn't trying to get anybody in trouble.
It's just that there's only one country in the "zee" world.
All the other jurisdictions with English as their/an official language
pronounce it "zed".
I believe the
>zPlane, zPlane ...
As I, and other Canadians, have already pointed out, this is a US-centric joke,
and only funny in the States.
Plane is just not funny in the other English Speaking countries.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
>Never met a paranoid Canadian before .
A conservative is a liberal who has been mugged!
As for paranoia, losing a job, or two, due to outsourcing will do it.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
Fo
>Not a big deal.
I'm not paranoid!
They are out to get us!
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with
>I have not work in mainframe environment and now I have to perform the
>functions of "System Programmer."
Sent under two different names:
John Smith and Hilario G.
Very suspicious to me.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
---
>JES3 on a local could have a different device than JES3 on the global.
You can do that with any multi-system config.
I thought we were talking about duplicates on the same system image.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
>Is this a JES3 system?
>I thought that JES3 allowed more than one volser online at a time with the
>same volser.
>I haven't worked with JES3 in a long time, so it may have changed.
Since it is MVS doing the complaining during the actual IPL, before the
remaining sub-systems start up, how can th
>Also, any system auto-operations product start-up script
That's why (imo) it's better to just list requirements.
Also, if you start listing methods, what happens if you haven't tested the
product, or the product changes?
Tend to your own knitting, and let the customer wear it.
-
I'm a SuperHe
>I still have to ask the question as to why is the IPL text release sensitive.
Because it is?
If that's the way IBM has designed it, live with it.
They've already documented, so what is the problem?
It is what it is.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
>Is it still preceived wisdom that the size of the PLPA data set should be
>minimal so that the LPA flows over to the COMMON data set?
>If so, what was/is the reason for that?
Back when DASD was expensive, it was common practice to save a spindle.
Today, some people say stop doing it because of
>>> "E" does no such thing.
>
>> Are we talking about E or $E?
>We are talking about SDSF action characters.
Well that was informative!
I usually overtype the service class, so I have NO idea what your response
meant!
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
---
>> It's also used to change service classes.
>"E" does no such thing.
Are we talking about E or $E?
E ,SRVCLASS=
Does exactly that.
Prior to Goal Mode, there was a PERFORM=nn
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
-
>
My IEFUSI (not yet updated for storage above the bar reasonably large defaults
are provided through SMFPRM00) makes REGION a binary switch. You either get my
generous defaults or you code REGION=0M and we give you everything can less a
buffer for recovery.
That's a very enlightened approach.
>Don't look at IEFUSI as "control", look at it as helping or setting a global
>option / default.
It is a control, when it doent let the user override it.
>The same can be said of many other things on the platform.
But, some things do not need to be controlled, and I've alway let the users
spe
>REGION is one of those things that should not be left to application JCL to
>manager properly.
I disagree (strongly) with that statement!
Tech Support is there to help, not control!
Yes. The config is a technical issue; the reason it's there is to support the
business, ie, appdev.
They know b
>That sounds useful but unless my memory is inaccurate an STC must be a single
>step
I believe your memory is inaccurate.
TSO must be a single step, or you get an abend - S622, iirc.
But, we've had multi-step CICS Regions as started tasks.
Clean-up; CICS; recovery.
PS: it's a started task (or
>Why are you so opposed to my mentioning some recommended approaches?
It's a matter what to do vs how to do it.
The recommendations tend to become instructions, and as SYSPROGs 'dumb down'
inconsistent implementations abound.
Just tell me what is needed, first.
If you think your recommendations
>So humor me, I'm not much of an operations person as you can tell, where would
>that be?
>One alternative is "your console automation system" but what are the
>recommended vanilla IBM alternatives?
>- A $VS command in the JESx initialization data set.
>- Where else?
None of those!
Just tell u
>I would prefer to say "put a start command in __ or wherever your shop
>usually puts start commands that are to execute after blah and blah are up.
You are still missing the point!
>You may use a console automation package to start FOO if your shop normally
>starts started tasks in this wa
>We are confused on which size to consider and whether we can add more space to
>our OSAM dataset which is at 150120 tracks now (on 3390 device).
Tracks ^= size of data.
There is an interrecord gap between each block.
So, the smaller the blocksize, in general, the smaller the amount of data on
>You want to use connect time to figure out MB/sec on FICON?
Never said that!
Just pointed out what EXCPs can/really do mean.
I'm well aware of the issues with FICON.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
-
>The EXCP*BLKSIZE=MB is not true for all access methods. I'd only assume this
>is correct for SAM-E, which is probably the majority of tape IO.
Ron, we've had this argument/discussion before.
With XA/ESA, or even slightly before, IOC was changed to allow either blocks or
8.3 ms of connect time.
>pressing PF7 (DOWN).
Picking a nit: F7 is UP; F8 is DOWN.
(Unless you've customised)
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email
>On my system, userid.sys.SPF3.LIST but cusomizable by the installation.
If that's the case, you can always go to ISPF options and change the
allocations size.
Make it large enough to not fail.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
-
>2) If your shop has a product like Mobius, $AVERS, XPTR, Dispatch, you can
>look at routing your syslog to them.
Last shop I worked at used SAR, or whatever it's called, now.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
---
>Which is why every mvs system (or VM, for that matter) can be IPL'd in just
>about any lpar (provided the necessary hardware is defined to that lpar).
We used to do that, at one place I worked, but, during the implementation of
GDPS, we lost that flexibility.
Our automation team claimed it was
>I don't think security policies should have any bearing on the TSO Region
>size. Just because it is something they can administer doesn't mean they
>should.
As I've already stated, that is my position, as a capacity analyst.
>But then, Elardus did have a smiley face on that sentence.
I caugh
>Since I'm more of a RACF guy, you may have difficulty to come up with a good
excuse... ;-D
Why are you so concerned about whether somebody should have a larger region
size, or not.
It's a hold-over from before memory became virtual and the small real memories
of the 1970's & 1980's.
With tod
>You'll have to migrate to VSE to get this feature.
Or have a front end to parse the data before submission.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access ins
>Why not just put them under WLM control, and then the system controls
how many you use (up to your set limits)? This is much easier than manually
stopping and starting initiators, and it takes the control away
from operators who usually do not understand the impact starting too many
initiators
>Wouldn't you be better served by putting the initiators under WLM control?
>Then you would have as many initiators as you need without over committing CPU.
1: WLM managed inits are not a panacea.
2: Undisciplined operators can still wreak havoc.
-
I'm a SuperHero with neither powers, nor motiv
>Its only for testing
Testing what?
If you're not going to put it in Production, what is the purpose?
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructio
>Funny how companies still figure on IBM to have all the answers when all they
>need to do is look internally.
>I've seen this more times than I can count.
Ditto.
I've lost my job twice due to outsourcing, and I'm pretty good at what I do.
Unfortunately, that makes you high-priced.
And, out-sourc
Sorry, this may be my ISP faul.
It was intended for the list, and I'm not sure which one.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
-Original Message-----
From: "Ted MacNEIL"
Date: Thu, 9 Sep 2010 22:55:21
To: Paul Gilmartin
Reply-To: eamacn...@
>Is there a JES2 display command that will return a job's job id?
$d'jobname'
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists.
>And for Ted's remark concerning ISPF SUBMIT vs. TSO, that's a detail of
>implementation that the user wanting an enhancement shouldn't need to be
>concerned with.
Possibly.
But, most used to know it.
It was documented as part of the ISPF user's guide, which also told you that
ISPF would reform
>Apparently IBM saw sufficient need or business case for the
facility to invest the resource to provide it in JES2.
>Why stop short of supporting it in ISPF?
It's actually TSO.
ISPF just manages the records to support the TSO/E submit restrictions.
-
I'm a SuperHero with neither powers, nor moti
>I got it.. zed Plain? Huh?
You mean: Eh?
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GE
>How do I search the archives of IBM-MAIN? I tried http://listserv.ua.edu (like
>http://listserv.uga.edu for the Assembler list) and http://bama.ua.edu but
>can't find the newslist server for this group.
One could probably start by reading the add on tacked on to the end of each
post:
-
I'm a
>For use on "Fantasy Island", there is no-frills machine called zPlain!
Which only makes sense in one country in the world.
In the US, it's pronounced 'zee'.
Everybody else pronounces it 'zed'.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
---
>This after numerous EMC marketting reps assured me, of a period of several
>months, that EMC had NEVER experienced a storage failure that resulted in ANY
>outage.
I KNOW of one in Canada in the mid-1990's.
What ever he's smoking, I'd like some.
-
I'm a SuperHero with neither powers, nor motiv
>I would absolutely LOVE a grass roots campaign to eliminate = as a token in
>any and all languages.
>Comparison should be ==. >Assignment should be :=.
In 1980, my final year at UOW, I wrote a paper stating exactly the same thing.
I got a good mark, and ended up with a double major (CS & Stats).
>I disagree, John. Spelling is VERY important in preserving the meaning of
>written communications.
I agree 100%!
When I was an elementary school student, in the 1960's, and I did a science
project, if I made a spelling, or grammatical, error, my science teacher would
take up to 5% off of the p
>These issues have been around for well over 40 years, and if they haven't been
>addressed then shame on whoever's responsible.
>It simply isn't acceptable to say that "something got messed up".
>This isn't 1975.
Unfortunately, everything old is new again.
People who don't learn from history ...
>You are confusing "subsystem" and "address space type". Under normal
>circumstances, TSUs also run under JES2 or JES3.
Yes, of course.
But, there is a definite sub-system called TSO, just as there is one called STC
(Started Task Control).
They all require JES, but they are separate sub-syst
>> A dunsel is still a dunsel.
>Thank you for expressing your minds using words out of dictionary
I'm allowed to use words that are common usage in my native tongue, aren't I?
Of course, only Star Trek fans would get the reference.
>There would be many applications for such feature. >Some batch
>Why don't you call it an exercise instead of a test, then you can always say
>that in an exercise you would expect something to fail and that's
the reason for the exercise.
If I were the keeper of the corporate buzzwords, I would.
I tried that route, at a major Canadian bank, and they told me i
>Isn't it obvious?
If it were obvious, would I ask why?
>More functionality is usually better.
A dunsel is still a dunsel.
>If you don't like it, then just
don't use it.
I never said I didn't like it!
I simply asked why.
I can see no use for the function, at the moment.
I wouldn't mind knowi
>Especially when every company that does DR testing always come back
>proclaiming how successful the test was.
Tell me about it!
I was involved, as a customer, when our service provider declared a test a
success when it didn't even meet more than 5 (out of almost 100) success
criteria.
Some app
> Would be nice to have(i.e TERM=TS from a batch job)
Why?
Wouldn't that violate so many integrity rules?
There is always the SEnd command.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN su
>the way I read the
statement from the JCL book, it says something totally different, and seemingly
totally wrong...
There's submitting and there's submitting.
Submit a job to get it to execute under JES2/3.
Submit text/commands, under TSO, and the output comes back to the terminal.
TERM=TS o
>Maybe the key is that the background job must still run under TSO? Is that
possible?
While you can run the TMP in the background, by definition, batch is
non-interactive, and runs under the JES (2 or 3) sub-system.
TERM=TS is for interactive work, under the TSO sub-system.
So, in short, it is
>I remember, in 1981, it was still called SPF.
The name changed in the mid 1980's, IIRC.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email
>No "foreign" CF can be
connected to the sysplex, even if physically connected via coupling links.
This is correct.
But, if you connect 'foreign' CF's up to a SYSPLEX, it will poll them, causing
performance problems on the 'PLEX where they are active.
Our hardware guys went connectivity/symmetr
>Don't you need also to specify SPACE on the DD statement? Or you
can use OLD and overwrite with a larger data set.
Yes you do.
Specify it with lots of commas.
It actually works.
You don't need to specify the primary, and you can specify secondaries, even
for datasets that had none.
The key is y
>CAPS ON is the very worst, because it so easily destroys data.
Only if you're not aware.
Any line you don't edit retains its case.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signo
>It is not preserved for future use of the dataset, even if the dataset is
>passed to the next step in the job.
But, and I experimented with this in the early 1990's -- at management's
request, you can use RLSE (with mod) and change the space allocation.
You can also add additional extents to a
>If you want to suppress this for all time, create an IMACRO to do this:
>NUM ON
>UNNUM
>NUM OFF
NOT a good idea if there is valid data in the columns which hold statement
numbers.
You should first check the mode, and just issue the NUM OFF.
-
I'm a SuperHero with neither powers, nor motivatio
>Does anybody know of a way to prevent the following from happening when
>setting numbers off?
I've done it with an initial EDIT MACRO.
Within that, if numbers are on, turn them off.
I forget the exact ISREDIT query/sub-command, but there is one that will tell
you about numbers.
-
I'm a SuperHe
>EMC told VITA that the outage was “unprecedented”
Precedent: A certain bank, in Toronto, had a similar failure many years ago
(mid-1990's).
Albeit, it was older technology, but the primary and secondary failed.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
>By the way, can you confirm that a structure will **never** be allocated in a
>CF not in the preference list (be it for initial allocation or for
>rebuilod/duplexing) ?
The documentation says that.
But, there is a way to test that:
a. Assume you have two CF's.
b. Let's call them CF01 & CF02.
c
>You forgot "Rin, Tin, Tin" and "Hoppalong Cassidy" :-) Not to mention. Roy
>Rogers, with Dale, Trigger, Buttermilk and a Jeep named "Nelly
Belle". :-)
My favourite was still ZORRO with the late Guy Williams, who also played John
Robinson (the father) in that 1960's laugher "Lost in Space".
>that this sort of warranty protects the performance guy at a customer site
>from actually having to understand IO performance.
Cheap shot
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscrib
>Um, don't the descendants of the 8086 -- you know, those pesky PCs -- have a
>stack implemented in hardware?
Don't DEC machines have them also?
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN sub
>Can anyone explain me why we are using Entry and Exit conventions in Mainframe
>assembler.
The reason for standards such as this one is so everybody knows what to expect.
We know that the called programme will save our status, restore it after we are
done, return to the next instruction after
>Very funny, but again: you
cite not mine words.
They may not be your words, but they are similar in intent.
>If you want to discuss
my opinions, then don't put third persons (and unrelated to the topic) opinions.
They are NOT unrelated.
>I said: the largest sysplex is bigger than any custom
My bad.
I forgot the 64 CP limit per LPAR when I responded.
Sorry.
But, I'm still curious about the largest SYSPLEX out there.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff /
>Parallel Sysplex can include
up to 32 systems, does that mean that Parallel Sysplex
now supports 32 x 80 = 2560 active CPs, or are we still
software bound to 2048 CPs per Parallel Sysplex?
Does anybody need that many, yet?
I'd love to know (academic interest) how big (CP's) the largest SYSPLEX i
>should be a very short interval indeed, one of some few hours, not one of days
>or weeks.
In your dreams!
We've been lucky to get one IPL of one LPAR a month.
>Toleration schemes, which attempt to make the non-incidental compresence of
>two versions of some component in the same MAS possibl
>In my judgment anyway, the presence of even two and certainly of three
>different levels of JES2 in the same MAS is 1) avoidable and 2) desirably
>avoided without compromising 24/7 availability, etc., etc.
I disagree.
Especially in a large shop, with multiple LPARS in the same SYSPLEX/MAS.
You
>Many businesses would not want to share business metrics e.g. number of
invoices with their suppliers.
That's possibly true.
But, invoices were just an example.
What would not be known is the number of line items, or the price per item.
>Pricing inputs on the value of outputs also seems wrong
>Per seat or machine capacity does not reflect the value to the buyer;
I disagree with that statement.
While I've not been a big fan of capacity based pricing (because the whole
model sucks), usage basis does make sense.
It's just a matter of what is your usage metric.
Seats, or transactions, ma
>Or has the practice of commenting code come into disrepute?
Real programmers don't document code.
It was hard to write, so it should be hard to read!
(8-{]}
A joke!
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
-
>Years ago, Dr Merrill stated that MXG probably processed more different date
>and time formats than any other software package.
MXG had that facilty mainly because SAS could do most of them.
But, once read, they were stored in internal (SAS) format.
Don't get me wrong.
MXG is a great example o
>SAS uses lots of date formats. ISO 8601 is a good spot to look for a large
>list.
Now, you have to be careful about that statement!
SAS displays a lot of formats.
But, usually, there is only one internal format.
Days from June 1, 1960, iirc.
-
I'm a SuperHero with neither powers, nor motivat
rame Discussion List
Date: Fri, 13 Aug 2010 13:45:05
To:
Reply-To: IBM Mainframe Discussion List
Subject: Re: share mainframe disk experience
We are in Rye NY ( Westchester County )
>Though I missed the start of these post's that's not true
we use EmC for our MF DASD w/ direct FICON - from Z/os Z/vm and Z/vse and have
had no problems w/ support
I guess it depends where you are.
One of my best friends was a STC (then STK, then Sun, then Oracle -- but he
left a long time ago
>The SMF30WID can be JES2, JES3, OMVS, STC, or TSO. MXG uses JOB for JES2 &
>JES2, and TSU for TSO.
Isn't there also a 6th sub-type for SYSTEM tasks, as a subset of STC?
We looked at it years ago, and we decided we didn't want to re-calculate all
our ratios.
-
I'm a SuperHero with neither pow
>This has been an interesting thread. It seems we all really ENJOY auditors...
> Can someone say prostate exam?
Bummer! (Sorry, couldn't resist).
But, there is a blood test (PSA) that will do the same thing).
In the last company I worked for, I convinced the powers that be to elongate
the tim
>For those doing chargeback, it has been recommended for quite a while that you
>move to "pricing bands" to account
for this variance.
When I took my first Capacity Planning course (1981), transaction-based
accounting was recommended, because (even then) CPU Time was variable.
I worked for a wh
>So if I want to get an operator fired, I'll just have my privileged task issue
>a $PJ1-, using that operator's console id?
1. What are you doing with a privileged task?
2. Don't sources of commands show up in SYSLOG?
3. Why would you do that?
4. Are you attempting to hijack this thread, or t
>If the consoles are in a secure room that should suffice all but the most
stringent security standards.
Unless you want to make your operators responsible for all commands issued from
their consoles.
I don't know if that would be considered stringent, but it is consistent with
the reasoning
>IIRC there is a statement in the SMF manual about CPU time not being
>repeatable.
This has been the case since pre-MVS/XA.
IBM documented this in many places over the last 25+ years, and gave
explanations at seminars such as CMG.
The simple explanation is, during one of the MVS/SP1.x releases,
>Our auditors are claiming this is "industry standard"
> and so we need to be doing it, even though our consoles are all behind locked
> doors.
At the risk of playing a tired tune, auditors don't set rules.
SME's do.
Auditors report.
Compliance officers enforce.
So, what do your SME's say?
(Su
>An increase of HSA could keep you from activating an LPAR, but that's all.
Since they are now independent, why?
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive acc
>On a Z10 it is
That's what I thought.
So, an increase in HSA would be irrelevent to paging.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send
>Why is it that you think you need to change the dataclass? Once the dataset
>has been allocated, changing the dataclass should have no effect that I can
>think of.
Data Class gets you DSCB attributes, but once the DSN is allocated, it's next
to meaningless.
-
I'm a SuperHero with neither pow
>I have a user which is used by many people on my z/os
That is a big security exposure.
Why not give each user their own ID?
That way, if something wrong (deliberate, or accident) is done, you know by
whom.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
---
>Larger HSA?
>What is your D M=STOR shows before & after the change?
Isn't the HSA independent of user memory, now?
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive acce
>The extended region size is the default value of 32 megabytes.
I've obviously misinterpreted that statement then.
I always read it as total.
But, the major point is that 16M does not give you all above.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
>Didn't R.11 allow IEFBR14 to issue HDELETEs as appropriate?
No, that's not quite accurate.
IEFBR14 hasn't changed.
The HDELETEs are done by some mechanism within the system/initiator.
Not to hijack this thread, but (as I've said before) that's a kludge!
-
I'm a SuperHero with neither powers, no
>Correction, if coding between 16M and 32M the default is 32M.
That's what I said:
16M above, is 32M in total.
-
I'm a SuperHero with neither powers, nor motivation!
Kimota!
--
For IBM-MAIN subscribe / signoff / archive access i
Sorry, I left a hanging fragment from the original post.
Re-sent, corrected.
>My understanding is that anything over 16MiB normally results in whatever
>below is available plus unlimited above.
Your understanding is wrong.
Anything over 16M does give you everything below, but it limits above.
>My understanding is that anything over 16MiB normally results in whatever
>below is available plus unlimited above.
Your understanding is wrong.
Anything over 16M does give you everything below, but it limits above.
If you specify between 16 and 32, you get 32 in total.
Whatever you specify abov
>YOU CANNOT TEST YOUR PRODUCTION before you move your production on the ONE
>VENDORs equipment.
Our last acquisition(s) we used time-finder, FDRPAS, or some other equivalent
to migrate.
And, we monitored while moving.
>To summarize, your approach is more or less "I don't care about any
technic
>> If the features are 'equal', regardless of implementation details, cost is
the
>> over-riding feature.
>[Ron Hawkins] I have no problem agreeing with that. >Your previous email
seemed to take a contrary position.
As I said, I may NOT have been clear, but when I reviewed my e-mails I thought
I
>...and you get what vendor wants. BTDT.
Again, I disagree.
>The performance strongly depends on workload type.
Absolutely.
>How do you define "my workload must be performed well and I must be happy of
>that"?
The easiest is to give them RMF data from your real Production workload.
>I saw (
I don't understand the uniqueness argument!
>One example that refutes the "DASD is Generic" argument is a multi-Petabyte
>customer that recently made their MIM Control file Cache Resident, which
>almost doubled the IO rate even though it was already 100% cache hit.
Our MIM control dataset, and
>I don't remember where I read or heard the story, but I think IBM preferred
>to use the term "storage" because "memory" implied that forgetting is
>possible. Therefore, to avoid conveying the impression that IBM computers
>could forget precious information -- or at least to suggest that forgetting
401 - 500 of 4516 matches
Mail list logo
