It is mentioned that making queries in Google Cloud Datastore is as expensive, with regards to time, as the number of results, which means that, for example, the time it takes to run any query would be proportional only to the number of matching results.
Can anyone give some explanation about how is it done in GCD or NoSQL Documented database?
I know there is a possible that you can implement the distributed system and run queries in parallel. But it is mentioned that the Datastore uses indexing to accomplish this, how would the indexing be in this way?