Sync Time Performance & "Read Table Rows" multiple times when Add Row Operation undertaken

Hi - I recently moved my company App from google sheets / Drive location to an excel workbook on the company's SharePoint. In doing so I noticed the sync times tripled or more and looking deeper in the Performance Analyzer I noticed it appears to read the same table 3 times...would appreciate if someone could give me a heads up on what is happening here and how i might fix my terrible sync issues? Many Thanks

Robert_Tooze_0-1692096961570.png

Robert_Tooze_1-1692097821179.png

 

 

0 6 353
6 REPLIES 6

Any chance that you have multiple virtual columns reading values from this same table more than once?

I do have a lot of virtual columns and maybe i need to just rebuild this (built it years ago fumbling my way through for a few months and its just cruised through great no troubles for the last 4 years - though slow to sync)...is it the Select formulas? i see there are 3 instances in the analyzer but i think i only have one select formula?? And is the virtual columns only a drain on sync time....are they also an issue on writing back to the spreadsheet?

And in my head I thought if it is writing to the spreadsheet(adding a row on the save action) it would only access the one table that holds all the records? and not some of the other ref tables...hhmmm what am i missing here?

Robert_Tooze_0-1692098611488.png

 

Every time when the app is syncing, it recalculates all virtual columns. If you are reading values a lot from different tables, it affects your sync time. It doesn't matter if it's a SELECT(), MAXROW(), LOOKUP() etc.

Ok thanks - I am understanding the sync side...but not understanding the 'Operation' of "Add Row" where you save the new row back to the sheet and how I have 3 instances of 'read table rows' pop up - is it possibly a bot/action? Looking deeper - I do have 2 actions that clears data out of 2 columns if the user changes one of the other column inputs (that turns off other columns in the row) - not sure if there is a better way to clear data if the user goes back and changes key columns? Thanks for the feedback

Robert_Tooze_1-1692101336192.png

 

Robert_Tooze_0-1692101034748.png

 

Those could explain it. Though it's easy to test if you disable them temporarely.

Ok I think I am getting it...I turned the actions off to clear the columns and that pulls me down to only one read table row...woo hoo - just knocked out 30 seconds of sync time...thanks heaps....now just need to work on the other 38 seconds -  a more detailed view tells me i have 16 seconds related to MicrosoftProvider_WriteTable (not sure what i can do about that??) and one virtual column running 15 sec which is trying to get the last time entry for a previous log so as to default it as the next log's start time - is there a better way to do this:

I use a virtual column "lastrow" =  maxrow(Tamping Logsheet, _RowNumber,[_ThisRow].[Shift Key1]=[Shift Key1])

then set the initial value for [Start Time] =[lastrow].[FINISH TIME]

Robert_Tooze_0-1692102230208.png

Robert_Tooze_1-1692102521521.png

Robert_Tooze_2-1692103550146.png

 

 

Top Labels in this Space