def~query-type

Definition

Controls how def~tables are exposed to queries

Hudi supports the following type of queries of stored data

• def~read-optimized-query

  Queries see the latest snapshot of the def~table as of a given commit / compaction def~instant-action; exposes only the base / columnar files in latest file slices to the queries and guarantees the same columnar query performance compared to a non-hudi columnar def~table.

  Following table summarizes the trade-offs between the different def~query-types.

  Trade-off	def~read-optimized-query	def~snapshot-query
  Data Latency	Higher	Lower
  Query Latency	Lower (raw base / columnar file performance)	Higher (merge  base / columnar file+ row based delta/log files)

• def~incremental-query

  Queries only see new records written to the def~table, since a given commit /delta-commit def~instant-action; effectively provides change streams to enable incremental data pipelines.
• def~snapshot-query

  Queries see the latest snapshot of def~table as of a given delta commit  or commit  def~instant-action.; In case of def~merge-on-read (MOR) table, it provides near-real time def~tables (few mins) by merging the base and delta files of the latest file slice on-the-fly. For def~copy-on-write (COW), it provides a drop-in replacement for existing parquet tables (or tables of same def~base-file type), while providing upsert/delete and other write side features.

Following table summarizes the trade-offs between the different def~query-types.

Related concepts

1. def~read-optimized-query
2. def~incremental-query
3. def~snapshot-query
4. def~timeline
5. def~table
6. def~commit
7. def~table-type