I am using private Vespa server. Given a schema with large number of documents, what is the best way to retrieve all document IDs? A query to match the relevant docs would be something like: "select id from my_container where my_field contains \"my_value\";" Indexing for the relevant field is enabled. The output size is limited by maxHits value which configured on server side. One also can use visit method with selection=my_container.my_field==\"my_value\". This works, but slow. In my case, sometimes, only small fraction of documents match search criteria and using visit seems to be inefficient. Continue reading...
