When you say the data in the files keeps changing, how much is changed?
You could save a lot of time by doing insert and updates for new and changed
records, instead of a truncate and insert for all records.
We run a similar nightly processes and we cut the copy time from 5 hours
down to 2. This is without having to drop and re-create indexes as well.
--
Thanks
Phill Carter
#562421
"determine" <anonymous (AT) discussions (DOT) microsoft.com> wrote
Quote:
Hi,
I have 42 text files. Records in each table are from 25,000 - 3.5
million. Data in these text files keep changing everyday. I have 42 tables
|
(set of temp and base) on sql server database. I run DTS jobs to load the
data from these text files into the temp tables. My temp tables doesn't
have the indexes. After the load is complete I truncate the data in the
base table and load the new data from temp table.
Quote:
There has been lot of performance issues. Some jobs complete fast and
couple of them take about 10 hours.
Is there an effective way of doing this. I want to redo the whole system.
I'm on SQL Server 2000/service pack 4 on windows NT operating system. Could
|
it be hardware problem. Do I need to upgrade the server.
Quote:
Please help.
Thanks a lot. |