July 19, 2016 at 4:40 pm
What's the best processing approach for individual SSAS Tabular tables? Here's the scenario.
I have two fact tables. They need to be updated every 10 minutes. One fact table has 2,006,325 rows. The other fact table has 361,551 rows.
I also have two dimension tables. They only need to be updated once a day. One is a date dimension with 4,046 rows. The other dimension table contains a hierarchy of our products. That table has 6,520,934 rows.
I want to have 4 separate jobs, one for each table. What is the best approach for the fact tables (every 10 minutes) vs. the dimension tables (once a day)?
August 20, 2016 at 10:23 am
For large fact tables, the best approach is to create partitions which can then be processed individually. This is especially helpful in the cases where you have to process regularly.
Even though you could partition any tables within a tabular model, it may not be possible to process individual partitions for dimension as any record can usually change at any point...unless that is not the case in your data model.
See the following reference: https://msdn.microsoft.com/en-us/library/hh230976.aspx
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply