| You might get a better response in TALLIS::TOPS which many
of us old(?!) 36-bitters still follow. While I have more than
a small amount of interest/expertise in all three of the
things you mentioned, I would NOT recommend trying to use
an emulator for such a situation. You are talking about
a database system(1022) which is VERY old (replaced by 1032
many years ago) together with all its tools (a Fortran
preprocessor and others) AND some assembly code thrown in
for good measure. Trying to move (and maintain) such a
beast is almost unthinkable.
I'd encourage them to start over with a modern data
base system rather than attempting to move the old one
via emulation. In the long run, this is the only sane
thing to do - other than to just KEEP the KS10 or SC40
running for as long as it meets their needs. Heck, that's
what I'd probably do. When do they retire?!
Cheers!
Dave Eklund
PS Can you tell us who the customer is (possibly off line)?
Dave E
|
| Jim --
Although I have a special fondness for -10s (and System 1022 even), I agree
with Dave Eklund -- they should put their effort into porting their
applications.
That is, unless their current hardware is failing, and won't last that long.
Then the emulator becomes more viable as a temporary measure. Talk to Andy
Riebs about it ([email protected], SOUSA::RIEBS, DTN 227-4044) for
full info, but the short story is: DEC decided not to market it, and
it's been handed back to its orginator, Ken Harrenstien ([email protected])
for distribution. It exists in both KS and KL variants. A while back,
there was a machine up on DEC's internal network, running TOPS-10 on the
emulator, so I know it supports TOPS-10.
Although XKL's TOAD-10 is a KL surrogate, it should still be able to run
user-mode applications that are running on a KS -- only the operating system
would have to be replaced. If the customer has a customized operating system,
that could be a problem. XKL currently supports TOPS-20, and is now working
on supporting TOPS-10, so that may not be ready yet. Also, the TOAD-1 would
be rather expensive unless they intended to keep running their current
application -- it probably wouldn't be cost-effective merely as a stopgap.
But back to porting the applications... It sounds like they no longer have
anyone in-house who understands the innards of their applications. Rather
than try to understand them in detail, what they might be better off doing
is to characterize what the applications *do*, then produce a new database
application that, from the outside, does what they want, then dump their
1022 data into a flat file, and reload it into the new application.
That would mean they'd "merely" need someone to help them extract the
data. Is it allowed for me to suggest a third party? The place I useta
work, the Locke Computer Center at the University of Washington,
decommissioned their -10 a couple of years ago, but their 1022 experts are
still there (now doing 1032, Ingres, etc.). The person to contact would
be Gerard Pence ([email protected], (206)543-9275).
Re. reply .1 posted by GIAMEM::NSULLIVAN -- does "He called the right guy
already" mean that the customer contacted *someone else*, not mentioned,
or that *Jim* was the right person? ;-) I'm always interested in hearing
about other -10 fans out there...
-- Pat
|