Don't miss your chance to take the Fabric Data Engineer (DP-600) exam for FREE! Find out how by attending the DP-600 session on April 23rd (pacific time), live or on-demand.
Learn moreNext up in the FabCon + SQLCon recap series: The roadmap for Microsoft SQL and Maximizing Developer experiences in Fabric. All sessions are available on-demand after the live show. Register now
Hi,
I'm trying to use the Incremental Refresh functionality in PBI Dataflows by connecting to 3 different tables via ODATA feed. All these tables are part of the same dataflow.
If I refresh the dataflow without using Incremental refresh, everything is refreshed in about 18 minutes. I'm trying to improve the performance as much as I can, so I enabled the Incremental Refresh. The initial refresh now fails with Too Many Request error message.
Just for test purposes, I split each table in a separate dataflow, and I noticed that one of them successfully completed, but the other 2 still fail. Below are the tables and # of rows.
Even though there is no big difference between Table 2 and Tables 3, Table 2 fails all the time when trying to read the same batch of data.
Entities | # of Rows | Incremental refresh status |
Table 1 | 902,960 | 429 – too many requests |
Table 2 | 202,130 | 429 – too many requests |
Table 3 | 171,890 | Completed in about 5 seconds |
Since the refresh works when not using incremental refresh, do you know what can cause getting Too Many Requests error message? Do you know what's the mechanism behind when using and when not using incremental refresh?
Thanks,
Teodora
What do you have for store and refresh parameters? Incremental refresh will make a call for each date granularity in the refresh range--so if you have 10 days set, it'll make 10 calls, if you have 2 months, it'll make two--if you have 60 days, it'll make 60 calls. Theses may be executing concurrently.
can you move the queries to separate dataflows?
Hi Jeff,
I moved each table in a separate dataflow and noticed that one of them successfully completed. Even though there is no big size difference between Tables 2 and Table 3, Table 2 fails.
The parameters I currently set are to bring data from the last 10 years and the granularity is 1 day. I think that might be one of the issues, at least for Table 1. I will give it a try to store data only for the last 2-3 years and the granularity to be weekly. Will come back with the end result, thanks for your help.
In my example here, initial refresh will query last 5 full years (so starting 2015-01-01), each refresh afterwards will clear out the previous 3 days and make three calls to refresh for each day.
DateTime column should be the create date for a record--don't use modified date. Incremental refresh is for appending new records, the only records updated will be those in the "refresh rows" range (since that data is cleared out and requeried).
I'm using Creation Date, I know that if I use Update Date I will start having duplicates.
This is the current configuration, the initital refresh is the one that fails. If I don't use incremental refresh, the initial refresh completes in about 18 minutes.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.
A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.
Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.
| User | Count |
|---|---|
| 48 | |
| 45 | |
| 41 | |
| 20 | |
| 17 |
| User | Count |
|---|---|
| 69 | |
| 64 | |
| 32 | |
| 31 | |
| 27 |