Is there a way to speed up reading records?

[jspur2003]

Ok our DB is alittle tricky as to it wont directly integrate with crystal so we exported the different database tables into files that crystal recognizes so we have 2 files that we need to work together Our main file which has all the sales and invoices ect has like 500,000 records and growing and the other file has customer info like Name, billing address, ect. The common link is Cust# so i can connect the 2 table the problem we are running into is it takes like 2 hrs to run a report and all we need right now is to get the Customer name.. My question is how can we speed this process up??? Any help is greatly appreciated...

 

[Pete56]

Put it into a data structure (like Access) where there is an index on the customer number field.

 

[synapsevampire]

Try posting:

Crystal version

Database/connectivity used - You might even post the database that you could not pull data from using Crystal as someone here might have a solution, this place is loaded with bright folk (BTW, that's not called integration). But definitely mention the type of tables you're reading from now...

A common solution would be to Import the data into a database, such as MS Access, index columns accordingly, build the query in Access, and then use the query as the data source for Crystal.

-k

 

[jspur2003]

Sorry our database uses ASNA Database Manager or Acceler8d DB Manager program and our database actual database is written by PFM and what we have done so far is the 2 files that we need to make our reports we exported them .txt file they are and we just named them Lot.txt and and Customer Master.txt for ease of use and we have had the best luck getting it to work with crystal... We are using Crystal 8.5 for the report writer... Now we havent tried loading the txt file into access and saving it as an access file to use in crystal yet We could go that route if necessary but if there are any other ways help would be appreciated... Thank you

 

[synapsevampire]

Crystal isn't a very good database engine, hence the performance.

That's why I suggested Access, or any real database (if one can call Access a real database without getting slapped down by the big kid database folk).

And importing is simple.

There is an option under Report called Report Bursting Indexes which might help with some performance concerns, but a 500K row table is fairly sizable.

-k

 

[jspur2003]

How many records will Access support? If it can hold a sizeable amount than its probably our viable solution