Simon Mara wrote:
Colleagues,
What frightens me is that I've seen up to seven [7] diallers on a small
server. I've had recent contact with a practice with 5, all working
somehow, believe it or not. All the usual suspects mentioned in the
discussion were present. For how long is the question, and what happens
when one or more of them break? What else will come down?
To me this complexity and overhead is the real problem.
Hi Greg,
When you say 7, do you mean running concurrently or simply installed?
Some use RAM an CPU speca all the time, some burst into life using the
task schedulers in MD and Windows.
The downloaders I've come across all have "run-connect-download-quit"
functionality, so they only steal resources while running. Typically this
would be only for a minute per program every two hours. I tend to avoid if
possible the downloaders built in scheduling that obviously requires CPU+RAM
resources to keep active. Cron, LaunchD or the built in Windows scheduler is
surely a better option than having X number of clients waiting around in
RAM? Are there downloaders out there that don't allow for this?
None of this is in contravention of the two decent points raised earlier:
1. Downloaders should be installed on client machines or a "sub-server".
2. A universal downloader would be ideal.
Yep, but you or I won't have much say over that as it's between the
practice and their path. providers.
I've only got a working knowledge of Genie, so wonder if other clinical
packages allow the client to insert the path result into the database from
the client (with the result residing on the client HD, not the server)? This
scenario means that there is no reliance on the client operating systems
connection to the server (mapped drives, UNC pathways etc). Downside is that
it is only practical for one computer to import the path, however this tends
to be a bigger limitation on paper than in reality.
The downloaders typically drop the results into a folder, wherever. My
understanding is that when MD users hit F4 or F5, from whichever PC, if
it knows where this folder is, it'll go to the folder and run the
importer locally that then drops the results into the MD database's
pathol.dbf table on the server, and if the results are in HL7 and are
diabetes-related, will put the coded atomic results into the
measures.dbf table. F5 also allows the GP to select their results for
viewing on their PC.
MD 2 isn't client-server. Each client pulls up the bits of the database
it wants, such as a patient's notes, over the network, and executes
changes in memory on the client PC, which then get written back to the
server where the data resides magnetically.
MD3 appears faster from a client, with it's sample database, because all
the processing is done on the server, and only commands and display of
resulting data goes across the LAN. MS SQL might be more efficient than
FoxPro as well, who knows.
Greg
--
Greg Twyford
Information Management & Technology Program Officer
Canterbury Division of General Practice
E-mail: [EMAIL PROTECTED]
Ph.: 02 9787 9033
Fax: 02 9787 9200
PRIVATE & CONFIDENTIAL
***********************************************************************
The information contained in this e-mail and their attached files,
including replies and forwarded copies, are confidential and intended
solely for the addressee(s) and may be legally privileged or prohibited
from disclosure and unauthorised use. If you are not the intended
recipient, any form of reproduction, dissemination, copying, disclosure,
modification, distribution and/or publication or any action taken or
omitted to be taken in reliance upon this message or its attachments is
prohibited.
All liability for viruses is excluded to the fullest extent permitted by
law.
***********************************************************************
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
