We use MB Foster ODBC.  I have used it with VB .Net and Cold Fusion and
it works very well.  The thru put is very good; it does not strain the
CPU.  I am a couple of versions behind (7.01) because I have to get all
the machines changed at one time and I have not yet found the time.  I
only use it to SELECT, UPDATE, INSERT into IMAGE databases and COGNOS
sub files.  I have never used it to access KSAM, Message File or FLAT
files.  We have processes that have been running for years using the
ODBC, with very little maintenance.

The ODBC are simple to load and just like accessing MS SQL database in
the coding.

We use a lot of FTP.  FTP works well for files.  I have struggled with
GUI FTP tools in MS SQL 2005 ETL's(Extract transform and load) and
BizTalk connections to certain machines.  The HP3000 is one; it is not
supported in the BizTalk list of Machines it connects.  If I could not
get the GUI tool to work I used FTP on the HP3000 or a batch job in
Windows as a work around.


Just some thoughts
Kent Wallace
Primary Health  

-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On
Behalf Of Michael Caplin
Sent: Thursday, January 03, 2008 11:52 AM
To: [log in to unmask]
Subject: [HP3000-L] PCLINK / ODBC

I'm looking for some help understanding/deciding how to proceed with a 
problem.  I'll try to be as brief as possible. 

System A is a windows box with an application running that responds to
the 
existence of files.  Ex: it sits waiting until FILEX shows up and then
does some 
processing and creates file Y then it goes back to waiting.


System C is an MPE box (ver 7).  There are applications running on it
that 
read/write to MSG files and flat files (no Ksam, Image, etc).  

The need is to move information from the MPE files on SYS C to/from the
files 
on SYS A.  At this time there's an application running on the 3000 that
waits 
for information to show up in certain MSG files.  When this happens, it
reads 
the file, writes it to a separate flat file and then runs PCLINK3
(reflection) as a 
son process to move this file to SYS A.  It then pauses and runs PCLINK3

again to get the new file created by the app on SYS A and moves the data
in 
it to the 3000.

This application ends up running PCLINK3 about 50,000 times in a 24 hour

period.  I'm looking into another way of doing this and am wondering if
the 
following has merit:

An application running on SYS B (a new box).  This app would have an
ODBC 
link to the 3000 to read and write files and an ODBC link to SYS A to
read and 
write files.

I have no experience with ODBC so I don't know if this could work or if
it would 
relieve the overhead of 50,000 runs of PCLINK every day.  Any thoughts
would 
be helpful as well as links to samples of code.

One last thing, there have been suggestions of replacing PCLINK with
FTP.  I 
don't know that this would be any better in fact I believe it would be
worse 
but thoughts on that are welcome as well.

TIA,
Mike

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *