Forum Discussion
Power Query Connecting to Datasource
It takes 2 minutes to connect to two Power Query tables in a 724 KB Excel file. The source data comes from two Excel files: the first Excel file is 10.3 MB and has five tables used in subsequent ...
Since this post I've also:
(1) Unlinked and Re-Linked OneDrive NO EFFECT
(2) Reset OneDrive NO EFFECT
(3) Put all files on Local C: Drive NO EFFECT
While on Local C: Drive:
(4) Removed the largest table (46,500 rows X 8 columns) that was also "unpivoted" NO EFFECT
(5) Removed all other non-essential tables NO EFFECT
(1) Unlinked and Re-Linked OneDrive NO EFFECT
(2) Reset OneDrive NO EFFECT
(3) Put all files on Local C: Drive NO EFFECT
While on Local C: Drive:
(4) Removed the largest table (46,500 rows X 8 columns) that was also "unpivoted" NO EFFECT
(5) Removed all other non-essential tables NO EFFECT
Query itself could be slow.
Excel table is internally quired two times, first one to analyse and define types, second one to rederive the data. There is the trick with M-scripting for workaround, but to use if that's really critical.
Sorting and FillUp/Down are quite expensive operations.
How do you do merging - Table.NestedJoin or Table.Join and how.
Other things could affect. Etc., etc. Chris Webb has series of posts related to PQ performance, one of them Chris Webb's BI Blog: Optimising The Performance Of Power Query Merges In Power BI, Part 5: Cross Joins . Perhaps here you could find something useful.
Such things as performance is too abstract to discuss without concrete file(s).
- Thank you Sergei. Much appreciated. These references were helpful. So were these:
https://learn.microsoft.com/en-us/power-bi/guidance/power-query-referenced-queries
https://learn.microsoft.com/en-us/power-query/new-dataflow-from-template