john230873
Programmer
The company I work for was bought up using Dataflex, migrated to Pervasive and is currently on Pervasive 2000i SP3. Due to how the company grew at the moment it has to stay with PSQL for the main transactions. The idea is to have a MSSQL server, load it once a night and use it as a data warehouse for reports. Currently the fastest way we have found to extract the data from Pervasive is using the API's, this takes about 3 hours (largest file being 4,000,000 records). To do this we have about 10 programs running at the same time which reads each record and write them to csv’s. Once all the CSV's are written then a DTS package runs on the MSSQL to load them in. Whole process takes about 6 hours.
I would like to do away with the CSV files and all the import programs and use the ODBC or OLEDB layer of Pervasive but I am having problems.
Speed is the big issue the importing reading of the PSQL data is so slow reading at about 400 records a second, is this the norm ?
I have noticed that the CPU on the computer running the server is at 100 %, this is a HT P4 2.8 gig. Should this happen?
Any help would be appreciated
I would like to do away with the CSV files and all the import programs and use the ODBC or OLEDB layer of Pervasive but I am having problems.
Speed is the big issue the importing reading of the PSQL data is so slow reading at about 400 records a second, is this the norm ?
I have noticed that the CPU on the computer running the server is at 100 %, this is a HT P4 2.8 gig. Should this happen?
Any help would be appreciated