View Running Queries
How to view and monitor running queries in StarRocks using current_queries and related views, analyze their resource consumption, and cancel long-running queries.
Gather statistics for CBO
This topic describes the basic concept of PhoenixAI CBO and how to collect statistics for the CBO. PhoenixAI introduces histograms to gather accurate data distribution statistics.
Use Bitmap for exact Count Distinct
This topic describes how to use bitmaps to compute the number of distinct values in PhoenixAI.
Bitmap indexing
This topic describes how to create and manage a bitmap index, along with usage cases.
SQL Plan Manager
How to use SQL Plan Manager in StarRocks to bind a fixed query plan to a SQL statement, preventing plan changes from statistics or data updates and stabilizing query performance.
Query Feedback
How to use Query Feedback in StarRocks to automatically analyze slow query plans, generate optimization guides via Query Plan Advisor, and improve subsequent query performance.
Synchronous materialized view
This topic describes how to create, use, and manage a synchronous materialized view (Rollup).
Use HLL for approximate count distinct
Background
Bloom filter indexing
This topic describes how to create and modify bloom filter indexes, along with how they works.
Asynchronous materialized views
This topic describes how to understand, create, use, and manage an asynchronous materialized view.
Asynchronous materialized views
6 items
Automatic Materialized View Recommendation
This topic describes how to leverage the Automatic Materialized View Recommendation feature to generate schemas for materialized views that can be used to accelerate the queries in your business scenarios.
Colocate Join
For shuffle join and broadcast join, if the join condition is met, the data rows of the two joining tables are merged into a single node to complete the join. Neither of these two join methods can avoid latency or overhead caused by data network transmission between nodes.
Caching
2 items
Use Lateral Join for column-to-row conversion
Column-to-row conversion is a common operation in ETL processing. Lateral is a special Join keyword that can associate a row with an internal subquery or table function. By using Lateral in conjunction with unnest(), you can expand one row into multiple rows. For more information, see unnest.
Computing distinct value count
2 items
Sorted streaming aggregate
How to use sorted streaming aggregate in StarRocks (available from v2.5) to optimize GROUP BY queries on pre-sorted data, reducing memory overhead compared to hash aggregation.
Accelerate COUNT(DISTINCT) and Joins with AUTO INCREMENT and Global Dictionary
How to use AUTO INCREMENT columns and Global Dictionary in StarRocks to replace string keys with integer IDs, accelerating COUNT(DISTINCT) and join operations.
Flat JSON
How to use Flat JSON in StarRocks to accelerate JSON queries by extracting common fields from JSON columns into flat columnar storage, reducing parsing overhead.
Accelerate Cardinality-Preserving Joins
How to use table pruning to accelerate cardinality-preserving joins in StarRocks by removing redundant joined tables that do not affect query results.
JIT Compilation for Expressions
How to enable and configure JIT (just-in-time) compilation for expressions in StarRocks to accelerate expression evaluation in query execution.
Skew Join V2
How to use Skew Join V2 in StarRocks to optimize JOIN queries with data skew by broadcasting skew values, improving performance for heavily skewed data distributions.