Hi Team
Hoping someone here can shed some light on the right way to approach this problem
I have a medium size data set that I need to calculate a rolling 13 week sum.
My data is stored in a SQL server table and to calculate I currently stream the data out and then process the calculation using the Multi-Row Fomula Tool.
The problem is Streaming the data out & then back in to write back to a new table causes a significant bottleneck on the workflow and it is taking about 7 hours to process 13 weeks of data.
Is anyone able to provide an example of an in-db fomula which can calculate a rolling sum (with some grouping options)
Any suggestions welcome
Data is structured similar to the below:
| Week | Product | Qty Sold |
| 16/08/2015 | 123 | 10 |
| 16/08/2015 | 555 | 5 |
| 23/08/2015 | 123 | 7 |
| 23/08/2015 | 555 | 78 |
| 30/08/2015 | 123 | 12 |
| 30/08/2015 | 444 | 3 |