Forum Discussion
Reducing data size - remove columns or merge with 2nd table size
Anonymous
I'm not using incremental refresh.
It's a recently created report. I created it in desktop with a weeks worth of data, now I've published it to service for 3 months worth of data and the file isn't loading (the circle spinning icon just goes for an hour+). I'm trying to use the manual refresh icon in Service "dataset + dataflows" tab.
Table A is a several .dat files on sharepoint appended in M query.
I've used this advice to create Table A, then use it as an internal source before merging with Table B in a seperate query
https://www.excelguru.ca/blog/2015/03/11/power-query-errors-please-rebuild-this-data-combination/
No network issues I'm aware of. It wasn't loading Friday when i first tried, and still isn't today.
- Anonymous6 years agoNot applicable
The dataset is currently connected to 3 .dat files (in csv format). each .dat file is a month worth of data and around 5-5.5gb each.
- Anonymous6 years agoNot applicable
Anonymous
So I've had a look in the detail and it is actually through up an error.
Table C is the result of merging Table A and Table B (Table A has "enable load" disabled is the PowerBI Desktop Query mode).
Data source error: {"error":{"code":"ModelRefresh_ShortMessage_ProcessingError","pbi.error":{"code":"ModelRefresh_ShortMessage_ProcessingError","parameters":{},"details":[{"code":"Message","detail":{"type":1,"value":"Unable to read data from the transport connection: An existing connection was forcibly closed by the remote host."}}],"exceptionCulprit":1}}} Table: TABLE_C. Cluster URI: WABI-NORTH-EUROPE-redirect.analysis.windows.net Activity ID: 9784f625-0fc4-4b01-ade0-530045abb358 Request ID: b151ec94-bd95-bf75-42ce-047262dca8de Time: 2020-05-22 18:42:26Z - nickyvv6 years ago
Most Valuable Professional
Hi Anonymous,
with importing files this big you might run into query timeouts.
An important question to me is: What is the source of your .dat files? Is it a database? Are you able to connect to that source directly, that would be much better. Depending on the source, you could even use the concept of Query Folding, which can potentially push the transformations back to the source system and give you much better performance and hence no issues with refreshing.