Re: Real-time updates with Dev/2000
Date: Sun, 3 Jan 1999 13:33:31 -0500
Message-ID: <76od0b$1rs$1_at_remarQ.com>
Would your user go for "same time"? I don't want to open a debate about what is "real-time" or exactly when is "now". If the user will go for some delay, you could use the application timer in the form to trigger queries. I work with SCADA (Supervisory Control and Data Acquistion) which is supposedly a real-time system. However, the field data is read once every two seconds, and goes into a memory resident database which is read by an operator interface every two seconds and displayed on a screen for the operator to review which means that my real-time data is at worst four seconds old. (Provided that the operator is awake)
The application timer would give you a single integrated solution. It could be a performance problem if you read the DB every two seconds, but it would allow the user to see the message counts updating. They would be old, but they would change. Is the object of the exercise to verify that the process that updates the message count is running, or is it to see the exact count at the exact instant? No matter what approach you take, there will be some delay between the real-time count and the displayed count. Your application will have to deal with network delays and message processing delays within the system. Is it worth the effort to get data that is milliseconds old over data that is seconds old?
I obviously don't know the details of your project, and I hope that the above may give you some thoughts for your user to consider.
Best of Luck,
Allan
John Coll <john_at_csoft.co.uk> wrote in message
news:76ahb1$ufu$1_at_news6.svr.pol.co.uk...
>Thanks for this helpful comment Frank. It looks worse than I thought! This
>really is interesting, I may have to abandon Oracle for the real-time part
>and have a second application for the dynamic stuff. Horrible from the user
>perspective. The user really wants one integrated appliaction able to
>access the database and see dynamic updates to "message counts" etc.
>
>thanks again for the input Frank.
>john
>
>
>
Received on Sun Jan 03 1999 - 19:33:31 CET