Tweet
Marc, We specify about 20-25 Files. We run the Job once a day in the evenings. Our Receivers, for a day's worth of activity add up to about 12 GIG. The Job runs in between 5-15 minutes. We've got a partitioned 825 running V5R2. Don't know what will happen in your case but why don't you run a test and see. If you dump all the transactions, you might be consuming lots of disk space, so be careful. Mike
-
-
DSPJRN Performance
We dump our Entries to a Type 3 Outfile. The reason being that the Date is in *ISO Format and easier to deal with than the Type 1 File. I think Type 1 gives you a Date in *Job Format but then neglects to tell you what the Job Date format was.... not too reliable. If for some reason you need to know the Receiver that contains the Entry you'll need to use the Type 5 Outfile. If you're going to specify the files, you'll probably want to also specify the Entry Specific Data Length. We make ours just slightly larger than the longest record length that we extract. I think you can let the DSPJRN Command calculate the length for you. But it seemed to me that it was calculating a length big enough to accomodate the biggest record in the Receiver Chain instead of the biggest record that we selected for extract. But that might just be superstition on my part. Mike -
DSPJRN Performance
I am using the DSPJRN command to create a file of journal records for a certain time period. I need this process to run as fast as possible. In doing this I am only interested in looking at the journals for a certain number of files. I wasn't sure if it would be quicker to use the DSPJRN to dump all the record level entries into a file and then filter out the records I don't care about through a HLL program. Or to use the DSPJRN to only dump the records of the files that I care about. Not sure if adding this layer of selection within the DSPJRN command would cause it to run significantly slower. Any thought would be appreciated. Thanks, Marc........Comment
Comment