Please keep in mind the following when executing any of the Torch programs:
- Increase available background jobs – The Node5 Architecture framework makes use of background work processes rather than dialog. Increase how many are available in the system using operation modes to ensure there is sufficient capacity available for parallel processing.
- Set the number of intervals appropriately – Each interval represents discrete and independent pieces of work to be processed. An example is that an interval represents equipment records with identifiers from 0000050000 through to 0000100000. If you intend to run the program with 10 parallel processes, then as a general rule of thumb is that you should have somewhere from 10 to 100 times as many intervals. Therefore, there should be at least 100 to 1000 intervals. A separate rule of thumb is that you take the number of objects to be processed (e.g. 1 million equipment records) and divide by 1000. In that case you would have 1000 intervals with 1000 equipment records within each interval.
- Monitoring Options – The progress of the program can be monitored through the use of the standard Node5 Architecture framework monitoring tool to understand progress. This is described in an earlier section of this document.
Thanks for your feedback.