Question:
I am having a hard time getting my OPO application to show decent performance. It’s running on a 32MB Win3.1 PC, and queries a Solaris x86 Oracle Server using TCP/IP as the network protocol. When I perform a query using the supplied query buttons, the query comes back fairly quick. But then when I want to issue a new query of the whole database, I click on the re-query button (the one that brings all the records back, so you’re not just further refining the first query, but querying the whole database). This time it takes forever to get the whole database back in order to perform a new query on the whole database.
Maybe my methodology is faulty. I’d love some guidance, because I’m ready to throw in the towel and forget about OPO.
Answer:
As I have said before in this space, because performance is an interaction ofmany different factors, it is virtually impossible to discuss it in anyspecific way in this forum.
I don’t understand your situation. Whenyou query a database, you send a SQL query to the database, the databaseprepares and executes the query, and then returns data. From theway you describe your situation, you are executing the same query that youdid initially, so I cannot understand why it would take longer. In fact, mostdatabases would return faster, since relational databases typically cachethe SQL statements recently used, and just re-use them if they come back again.
However, your use of the phrase ‘whole database’ gives me pause. I hopethat you are setting the RowFetchMode property to one of theincremental fetch settings. These settings, “Fetch As Needed” and “FetchCount First,” return only the first few rows from a query, so they end upgiving faster and more consistent response to a query.
Other than that, I can’t think of anything. To make sure that the problemis not in your database, I suggest testing this same query withanother tool, such as SQL*Plus for Oracle. If this has the sameperformance characteristic, look to your database, not OPO.
