Source: S3 Query Engine: Athena In our use case, several partitions (and hence, files) are added to S3 continuously and the partitions are made available immediately via dynamic partition projection. So the cost of adding partitions is not high.

Now, I am porting this to GCP GCS and Presto with Hive metastore (Trino distribution, to be exact). (I haven't tried DataProc yet, both Presto and Hive metastore will be self managed)

I am looking for something close to dynamic partition projection, the way it happens in Athena. However, I haven't found anything automatic. It is either manually adding partitions or repairing the entire table via the procedure system.sync_partition_metadata

Is there anything less expensive in Presto that'd help me in adding the partitions to the metastore without requiring a full scan?