Acessing Progress from ColdFusion

[dboulden]

Hi,
I'm a veteran ColdFusion developer, but completely new to Progress. I'm creating a web based system to retrieve info from a Progress v8.3B database using ColdFusionMX. I'm using the DataDirect (Merant) v3.7 ODBC driver to connect to the Progress datbase on a SCO Unix box, CF is installed on a Win2K machine using IIS web server. It all connects OK, and I'm getting data back, but it is horrendously slow (> 5 mins) when it should be near instantaneous. Are there any special requirements to be aware of when formulating the SQL calls that differs from the normal SQL servers? The type of call I'm doing is

SELECT name, address, postcode FROM account WHERE name LIKE "ABC%"

thanks in advance!

Dave Boulden.

 

[ChrisPaulson]

Have you checked index usage?

How many records?

Have you considered using webspeed, it might be easier in the long run?

 

[dboulden]

Haven't been able to check for indexes yet, but only around 20,000 records so I can't see that it is a linear search causing that much delay. It almost seems as though it is waiting for something to time-out before returning the data. Did some debug stats... compile and run the CF page took 93 ms, the query took 1 million ms. The same query run against a re-contructed copy of the data in a MySQL database with no indexes was almost instantaneous. I'm a bit baffled... might be some kind of option not set right in the ODBC dsn. Does Progress require any kind of terminator character on the SQL call? (like MySQL has a ';')

 

[ChrisPaulson]

20,000 records....

I'd check the index, Progress is not very fast!

 

[ChrisPaulson]

Correction :-

Progress through an ODBC driver is not fast. It is probably pushing all 20K records though the middleware.

 

[dboulden]

OK, thanks Chris. I'll have to try & find out what the indexes are (once I've sorted out VPN access to the customers site!)

Dave.

 

[jorgepino]

Coldfusion and Progress

it been hard to get those two working correctly but if you can use store procedures to pre filter your data may make it go faster