Hi All,
"Using the LoadDocument adds a loaded document to the tracking list. This may cause very expensive calculations to occur, especially when multiple documents are tracking the same document."
But how exactly does that affect indexing performance?
We have an index which has two calls to LoadDocument. One of those documents are rarely updated, whereas two of them are updated somewhat frequently. We had some issues with this index with errors like "Failed to allocate memory, 16GB already allocated to this thread". When we look at this index in the dashboard, the progress bars on the index update process shows all three documents, even though one of the collections are rarely updated. Does this have to do with the above message in the documentation?
How could we improve this index?
meshcloud GmbH
Sonnemannstr. 5
60314 Frankfurt am Main
Germany
Follow us
Registergericht: Amtsgericht Frankfurt am Main, HRB 116189
Geschäftsführer: Dr. Jörg Gottschlich