QVD, ODBC Datasource, and Incremental load

[#1]

Hi,

So, I have an app that connects to an ODBC datasource.  When I put this app on our access point and had the ODBC driver properly configured, the QMC is able to reload a task.  It takes about 10 minutes, which I don't think is very efficient.

So I am curious what you guys think is a more efficient route...should I consider a combination of creating a QVD + incremental load off the ODBC DB?  Should I simply consider incremental loads?  New data is added every day and this app can and probably will be utilized every day.

Just looking to reduce strain on the QMC and reduce runtime from 10 minutes to maybe a minute or so.

Thoughts?

[#2]

Incremental loads are nice because they pull in only the changed/new data.

Keep in mind that QVDs will increase load speed greatly, reducing space in memory by up to 90%.  You can convert the data you have into QVDs, then drop the tables you've loaded and reload the QVD version of the table.  Examples:

//**************SUBROUTINE DEFINITIONS************************

//drop original

Sub StoreAndDrop(vTableName)

Store [$(vTableName)] into [$(vTableName).qvd];

Drop Table [$(vTableName)];

End Sub

//keep original

Sub StoreAndKeep(vTableName)

Store [$(vTableName)] into [$(vTableName).qvd];

End Sub

So, you might decide to load the data as you get it (via csv or xls for example) then use the above examples to make QVDs in a separate "QVD maker" app.  Those QVDs can then be used in your main app instead of whatever your other data source is.