Jim Lawrence (AccessD)
accessd at shaw.ca
Sat May 3 10:01:33 CDT 2003
Hi Marcus: I will add my comment inline... -----Original Message----- From: accessd-bounces at databaseadvisors.com [mailto:accessd-bounces at databaseadvisors.com]On Behalf Of Marcus Tewksbury Sent: Saturday, May 03, 2003 7:14 AM To: accessd at databaseadvisors.com Subject: RE: [AccessD] Desperately Seeking! First Off, thank you to everyone for contributing your thoughts - it is very much appreciated. Let me see if I can account for this and reformulate... 1st.) The notion of deploying multiple copies of the .adp versus one centralized one to limit the # or recordsets is retarded. My line of thinking was one application = one recordset, but that can't be right, right? The application would create new recordsets for each user as they went. The number of recordsets created and therefore connections has nothing to do how many copies of the application being used. It depends how the application is written. Of course if you have set up your recordsets to all have persistent connections then you will quickly run out of recourses and bandwidth. so, I think my problem/solution deals more with the recordset maintenance itself. I'm going to try and break this down into over-writing, and viewing. Over-writing: Yeap - I had been using optimistic locks. I did a pretty good job of abstracting the code so it will be easy enough to change the records to pessimistic. I'm also going to play around with running "at the client" and "at the server" Viewing: Inside the app I want the users to be able to see deletions, additions, etc. made by other users. Which I presume means I am going to have to Refresh my recordsets. Is anyone familiar with ODBC refresh rates? Should I periodically issue Refresh methods on the recordset? How does a refresh compare to a requery performance wise? Because I know requerying just won't be acceptable because of recordset size. I have not used ODBC but performance will depend more on how much data has to be acquired from the BE. Keeping the recordsets small is one way. The refresh or resync method returns any changes in a specific record but the requery method re-executes the query that created the original recordset and therefore reflects any changes to the dataset. The resync method would generate an error if the current record had been deleted by another user. Again, thanks to everyone for your thoughts! - Marcus _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus _______________________________________________ AccessD mailing list AccessD at databaseadvisors.com http://databaseadvisors.com/mailman/listinfo/accessd Website: http://www.databaseadvisors.com