vinothchandar commented on code in PR #12409: URL: https://github.com/apache/hudi/pull/12409#discussion_r1867544457
########## website/docs/metadata_indexing.md: ########## @@ -10,18 +10,65 @@ The [pluggable indexing subsystem](https://www.onehouse.ai/blog/introducing-mult of Hudi depends on the metadata table. Different types of index, from `files` index for locating records efficiently to `column_stats` index for data skipping, are part of the metadata table. A fundamental tradeoff in any data system that supports indices is to balance the write throughput with index updates. A brute-force way is to lock out the writes -while indexing. However, very large tables can take hours to index. This is where Hudi's novel asynchronous metadata -indexing comes into play. +while indexing. Hudi supports index creation using SQL, Datasource as well as async indexing. However, very large tables Review Comment: please cover how concurrency control is done.. Just write out index building protocols as simple bullets. ########## website/docs/sql_ddl.md: ########## @@ -582,6 +597,28 @@ SHOW PARTITIONS hudi_table; --Drop partition： ALTER TABLE hudi_table DROP PARTITION (dt='2021-12-09', hh='10'); ``` + +### Show and drop index + +**Syntax** + +```sql +-- Show Indexes +SHOW INDEXES FROM tableIdentifier; Review Comment: please use consistent case (upper, lower, camel) as rest of docs -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
