Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Howdy,
I'm kind of confused about incremental load from log files...
Is it as easy as...
Buffer (Incremental)
Load *
From
O:\DavidC\3CX\*.txt
(txt, codepage is 1252, no labels, delimiter is ',', msq, filters(
Rotate(left)));
The question is, why is this not speeding up my reload times. I'm loading phone call log files, which grow daily. Shouldn't this script ONLY load the new files? Every day...my reaload time grows and grows as my number of log files.
-David
Yes, it is that easy.
You will see "loading" messages for the "old" files, but it is only checking the filesize and not actually loading any data from the files -- it gets that data form the buffered QVD.
What I'm not sure about is the effect of the filters(). Can you try it without the filter as a timing test?
-Rob
Do you mean my "rotate(left" cross table part of the script?
Yes, I meant remove the
filters(
Rotate(left));
I know it won't give you the correct result, but just as a test.
-Rob
OK, i'll update after the test.
What's the different in the script execution window between records with the "(buffered)" and the ones without?
-DC
I think you see a "buffered" message at the beginning of the load, but other than that the progress window looks the same.
-Rob