Hi,
I am trying to fetch data from Log Analytics from powerBI using Log analytics RestAPI.
The API has limitation of 500000 rows per query.
And hence I am thinking of handling large dataset at powerBI by calling multiple RestAPI (with different timerange).
I want to make this dynamic and hence I created function (which functionality code and call Rest API on Log analytics workspace).
Then I have a loop which basically calls this function iteratively for different time range and finally append the data.
let
list1 = List.Generate(()=>[Counter=90, table2 = null],
each [Counter] >= 0,
each [table2 = PIFunc([Counter], [Counter]-30),Counter=[Counter]-30],
each _[table2]),
// skip first null value
finalList = List.Skip(list1),
Result = Table.FromList(finalList, Splitter.SplitByNothing(), null, null),
#"Expanded Column1" = Table.ExpandTableColumn(Result, "Column1", {"BackupItemUniqueId", "ProtectedContainerUniqueIdData", "AsOnDateTime"}, {"BackupItemUniqueId", "ProtectedContainerUniqueIdData", "AsOnDateTime"})
in
#"Expanded Column1"
In the above code, it calls the function PIFunc for different time range (eg - 90 to 60 days, 60 to 30 days.. ).
Like PIFunc(90,60), PIFunc(60,30), PIFunc(30, 0).
Finally the table data will be in the list and then table is generated.
My objective is I would like to avoid any heavy operation (in terms of memory due to 1GB dataset limitation) performed at powerBI. That's why I am using LA to do all summarization and aggregation at LA.
I would like to know if it leads to high memory consumption or not ? Does it depend on intermediate table data or not?
If so, what is the right way to write some sort of loop to fetch small datasets iteratively and then combine data at powerBI in a single table?
