Feature Support Detail

Feature Support	TiSpark 2.4.x	TiSpark 2.5.x	TiSpark 3.0	TiSpark master
SQL select without tidb_catalog	✔	✔
SQL select with tidb_catalog		✔	✔	✔
DataFrame append	✔	✔	✔	✔
DataFrame reads	✔	✔	✔	✔
SQL show databases	✔	✔	✔	✔
SQL show tables	✔	✔	✔	✔
SQL auth		✔	✔	✔
SQL delete from with tidb_catalog			✔	✔
TLS			✔	✔
DataFrame auth				✔

Read

you can read with both SQL select and DataFrame reads

Push down

TiSpark can push down some operators to TiKV

• predicate
• aggregate

need to be described more clearly

Table Partition

TiSpark currently supports Range Partition and Hash Partition. Users can select data from the Range Partition table and the Hash Partition table through TiSpark.

In most cases, TiSpark use a full table scan on partition tables. Only in certain cases, TiSpark applies partition pruning. For more details, see here.

Expression Index

tidb-5.0 supports Expression Index.

TiSpark currently supports retrieving data from table with Expression Index, but the Expression Index will not be used by the planner of TiSpark.

TiFlash

TiSpark can read from TiFlash with the configuration spark.tispark.isolation_read_engines

Write

You can only write to TiDB with DataFrame append now

Not supported

• write to the partition table

Feature List

• Stale read
• TiSpark authorization
• TiSpark delete
• TiSpark TLS
• TiSpark Telemetry