- Docs Home
- About TiDB
- Quick Start
- Develop
- Overview
- Quick Start
- Build a TiDB Cluster in TiDB Cloud (Developer Tier)
- CRUD SQL in TiDB
- Build a Simple CRUD App with TiDB
- Example Applications
- Connect to TiDB
- Design Database Schema
- Write Data
- Read Data
- Transaction
- Optimize
- Troubleshoot
- Reference
- Cloud Native Development Environment
- Third-party Support
- Deploy
- Software and Hardware Requirements
- Environment Configuration Checklist
- Plan Cluster Topology
- Install and Start
- Verify Cluster Status
- Test Cluster Performance
- Migrate
- Overview
- Migration Tools
- Migration Scenarios
- Migrate from Aurora
- Migrate MySQL of Small Datasets
- Migrate MySQL of Large Datasets
- Migrate and Merge MySQL Shards of Small Datasets
- Migrate and Merge MySQL Shards of Large Datasets
- Migrate from CSV Files
- Migrate from SQL Files
- Migrate from One TiDB Cluster to Another TiDB Cluster
- Migrate from TiDB to MySQL-compatible Databases
- Advanced Migration
- Integrate
- Overview
- Integration Scenarios
- Maintain
- Monitor and Alert
- Troubleshoot
- TiDB Troubleshooting Map
- Identify Slow Queries
- Analyze Slow Queries
- SQL Diagnostics
- Identify Expensive Queries Using Top SQL
- Identify Expensive Queries Using Logs
- Statement Summary Tables
- Troubleshoot Hotspot Issues
- Troubleshoot Increased Read and Write Latency
- Save and Restore the On-Site Information of a Cluster
- Troubleshoot Cluster Setup
- Troubleshoot High Disk I/O Usage
- Troubleshoot Lock Conflicts
- Troubleshoot TiFlash
- Troubleshoot Write Conflicts in Optimistic Transactions
- Troubleshoot Inconsistency Between Data and Indexes
- Performance Tuning
- Tuning Guide
- Configuration Tuning
- System Tuning
- Software Tuning
- SQL Tuning
- Overview
- Understanding the Query Execution Plan
- SQL Optimization Process
- Overview
- Logic Optimization
- Physical Optimization
- Prepare Execution Plan Cache
- Control Execution Plans
- Tutorials
- TiDB Tools
- Overview
- Use Cases
- Download
- TiUP
- Documentation Map
- Overview
- Terminology and Concepts
- Manage TiUP Components
- FAQ
- Troubleshooting Guide
- Command Reference
- Overview
- TiUP Commands
- TiUP Cluster Commands
- Overview
- tiup cluster audit
- tiup cluster check
- tiup cluster clean
- tiup cluster deploy
- tiup cluster destroy
- tiup cluster disable
- tiup cluster display
- tiup cluster edit-config
- tiup cluster enable
- tiup cluster help
- tiup cluster import
- tiup cluster list
- tiup cluster patch
- tiup cluster prune
- tiup cluster reload
- tiup cluster rename
- tiup cluster replay
- tiup cluster restart
- tiup cluster scale-in
- tiup cluster scale-out
- tiup cluster start
- tiup cluster stop
- tiup cluster template
- tiup cluster upgrade
- TiUP DM Commands
- Overview
- tiup dm audit
- tiup dm deploy
- tiup dm destroy
- tiup dm disable
- tiup dm display
- tiup dm edit-config
- tiup dm enable
- tiup dm help
- tiup dm import
- tiup dm list
- tiup dm patch
- tiup dm prune
- tiup dm reload
- tiup dm replay
- tiup dm restart
- tiup dm scale-in
- tiup dm scale-out
- tiup dm start
- tiup dm stop
- tiup dm template
- tiup dm upgrade
- TiDB Cluster Topology Reference
- DM Cluster Topology Reference
- Mirror Reference Guide
- TiUP Components
- PingCAP Clinic Diagnostic Service
- TiDB Operator
- Dumpling
- TiDB Lightning
- TiDB Data Migration
- About TiDB Data Migration
- Architecture
- Quick Start
- Deploy a DM cluster
- Tutorials
- Advanced Tutorials
- Maintain
- Cluster Upgrade
- Tools
- Performance Tuning
- Manage Data Sources
- Manage Tasks
- Export and Import Data Sources and Task Configurations of Clusters
- Handle Alerts
- Daily Check
- Reference
- Architecture
- Command Line
- Configuration Files
- OpenAPI
- Compatibility Catalog
- Secure
- Monitoring and Alerts
- Error Codes
- Glossary
- Example
- Troubleshoot
- Release Notes
- Backup & Restore (BR)
- Point-in-Time Recovery
- TiDB Binlog
- TiCDC
- Dumpling
- sync-diff-inspector
- TiSpark
- Reference
- Cluster Architecture
- Key Monitoring Metrics
- Secure
- Privileges
- SQL
- SQL Language Structure and Syntax
- SQL Statements
ADD COLUMN
ADD INDEX
ADMIN
ADMIN CANCEL DDL
ADMIN CHECKSUM TABLE
ADMIN CHECK [TABLE|INDEX]
ADMIN SHOW DDL [JOBS|QUERIES]
ADMIN SHOW TELEMETRY
ALTER DATABASE
ALTER INDEX
ALTER INSTANCE
ALTER PLACEMENT POLICY
ALTER TABLE
ALTER TABLE COMPACT
ALTER TABLE SET TIFLASH MODE
ALTER USER
ANALYZE TABLE
BACKUP
BATCH
BEGIN
CHANGE COLUMN
COMMIT
CHANGE DRAINER
CHANGE PUMP
CREATE [GLOBAL|SESSION] BINDING
CREATE DATABASE
CREATE INDEX
CREATE PLACEMENT POLICY
CREATE ROLE
CREATE SEQUENCE
CREATE TABLE LIKE
CREATE TABLE
CREATE USER
CREATE VIEW
DEALLOCATE
DELETE
DESC
DESCRIBE
DO
DROP [GLOBAL|SESSION] BINDING
DROP COLUMN
DROP DATABASE
DROP INDEX
DROP PLACEMENT POLICY
DROP ROLE
DROP SEQUENCE
DROP STATS
DROP TABLE
DROP USER
DROP VIEW
EXECUTE
EXPLAIN ANALYZE
EXPLAIN
FLASHBACK TABLE
FLUSH PRIVILEGES
FLUSH STATUS
FLUSH TABLES
GRANT <privileges>
GRANT <role>
INSERT
KILL [TIDB]
LOAD DATA
LOAD STATS
MODIFY COLUMN
PREPARE
RECOVER TABLE
RENAME INDEX
RENAME TABLE
REPLACE
RESTORE
REVOKE <privileges>
REVOKE <role>
ROLLBACK
SAVEPOINT
SELECT
SET DEFAULT ROLE
SET [NAMES|CHARACTER SET]
SET PASSWORD
SET ROLE
SET TRANSACTION
SET [GLOBAL|SESSION] <variable>
SHOW ANALYZE STATUS
SHOW [BACKUPS|RESTORES]
SHOW [GLOBAL|SESSION] BINDINGS
SHOW BUILTINS
SHOW CHARACTER SET
SHOW COLLATION
SHOW [FULL] COLUMNS FROM
SHOW CONFIG
SHOW CREATE PLACEMENT POLICY
SHOW CREATE SEQUENCE
SHOW CREATE TABLE
SHOW CREATE USER
SHOW DATABASES
SHOW DRAINER STATUS
SHOW ENGINES
SHOW ERRORS
SHOW [FULL] FIELDS FROM
SHOW GRANTS
SHOW INDEX [FROM|IN]
SHOW INDEXES [FROM|IN]
SHOW KEYS [FROM|IN]
SHOW MASTER STATUS
SHOW PLACEMENT
SHOW PLACEMENT FOR
SHOW PLACEMENT LABELS
SHOW PLUGINS
SHOW PRIVILEGES
SHOW [FULL] PROCESSSLIST
SHOW PROFILES
SHOW PUMP STATUS
SHOW SCHEMAS
SHOW STATS_HEALTHY
SHOW STATS_HISTOGRAMS
SHOW STATS_META
SHOW STATUS
SHOW TABLE NEXT_ROW_ID
SHOW TABLE REGIONS
SHOW TABLE STATUS
SHOW [FULL] TABLES
SHOW [GLOBAL|SESSION] VARIABLES
SHOW WARNINGS
SHUTDOWN
SPLIT REGION
START TRANSACTION
TABLE
TRACE
TRUNCATE
UPDATE
USE
WITH
- Data Types
- Functions and Operators
- Overview
- Type Conversion in Expression Evaluation
- Operators
- Control Flow Functions
- String Functions
- Numeric Functions and Operators
- Date and Time Functions
- Bit Functions and Operators
- Cast Functions and Operators
- Encryption and Compression Functions
- Locking Functions
- Information Functions
- JSON Functions
- Aggregate (GROUP BY) Functions
- Window Functions
- Miscellaneous Functions
- Precision Math
- Set Operations
- List of Expressions for Pushdown
- TiDB Specific Functions
- Clustered Indexes
- Constraints
- Generated Columns
- SQL Mode
- Table Attributes
- Transactions
- Garbage Collection (GC)
- Views
- Partitioning
- Temporary Tables
- Cached Tables
- Character Set and Collation
- Placement Rules in SQL
- System Tables
mysql
- INFORMATION_SCHEMA
- Overview
ANALYZE_STATUS
CLIENT_ERRORS_SUMMARY_BY_HOST
CLIENT_ERRORS_SUMMARY_BY_USER
CLIENT_ERRORS_SUMMARY_GLOBAL
CHARACTER_SETS
CLUSTER_CONFIG
CLUSTER_HARDWARE
CLUSTER_INFO
CLUSTER_LOAD
CLUSTER_LOG
CLUSTER_SYSTEMINFO
COLLATIONS
COLLATION_CHARACTER_SET_APPLICABILITY
COLUMNS
DATA_LOCK_WAITS
DDL_JOBS
DEADLOCKS
ENGINES
INSPECTION_RESULT
INSPECTION_RULES
INSPECTION_SUMMARY
KEY_COLUMN_USAGE
METRICS_SUMMARY
METRICS_TABLES
PARTITIONS
PLACEMENT_POLICIES
PROCESSLIST
REFERENTIAL_CONSTRAINTS
SCHEMATA
SEQUENCES
SESSION_VARIABLES
SLOW_QUERY
STATISTICS
TABLES
TABLE_CONSTRAINTS
TABLE_STORAGE_STATS
TIDB_HOT_REGIONS
TIDB_HOT_REGIONS_HISTORY
TIDB_INDEXES
TIDB_SERVERS_INFO
TIDB_TRX
TIFLASH_REPLICA
TIKV_REGION_PEERS
TIKV_REGION_STATUS
TIKV_STORE_STATUS
USER_PRIVILEGES
VARIABLES_INFO
VIEWS
METRICS_SCHEMA
- UI
- TiDB Dashboard
- Overview
- Maintain
- Access
- Overview Page
- Cluster Info Page
- Top SQL Page
- Key Visualizer Page
- Metrics Relation Graph
- SQL Statements Analysis
- Slow Queries Page
- Cluster Diagnostics
- Monitoring Page
- Search Logs Page
- Instance Profiling
- Session Management and Configuration
- FAQ
- CLI
- Command Line Flags
- Configuration File Parameters
- System Variables
- Storage Engines
- Telemetry
- Errors Codes
- Table Filter
- Schedule Replicas by Topology Labels
- FAQs
- Release Notes
- All Releases
- Release Timeline
- TiDB Versioning
- TiDB Installation Packages
- v6.2
- v6.1
- v6.0
- v5.4
- v5.3
- v5.2
- v5.1
- v5.0
- v4.0
- v3.1
- v3.0
- v2.1
- v2.0
- v1.0
- Glossary
Explain Statements in the MPP Mode
TiDB supports using the MPP mode to execute queries. In the MPP mode, the TiDB optimizer generates execution plans for MPP. Note that the MPP mode is only available for tables that have replicas on TiFlash.
The examples in this document are based on the following sample data:
CREATE TABLE t1 (id int, value int);
INSERT INTO t1 values(1,2),(2,3),(1,3);
ALTER TABLE t1 set tiflash replica 1;
ANALYZE TABLE t1;
SET tidb_allow_mpp = 1;
MPP query fragments and MPP tasks
In the MPP mode, a query is logically sliced into multiple query fragments. Take the following statement as an example:
EXPLAIN SELECT COUNT(*) FROM t1 GROUP BY id;
This query is divided into two fragments in the MPP mode. One for the first-stage aggregation and the other for the second-stage aggregation, also the final aggregation. When this query is executed, each query fragment is instantiated into one or more MPP tasks.
Exchange operators
ExchangeReceiver
and ExchangeSender
are two exchange operators specific for MPP execution plans. The ExchangeReceiver
operator reads data from downstream query fragments and the ExchangeSender
operator sends data from downstream query fragments to upstream query fragments. In the MPP mode, the root operator of each MPP query fragment is ExchangeSender
, meaning that query fragments are delimited by the ExchangeSender
operator.
The following is a simple MPP execution plan:
EXPLAIN SELECT COUNT(*) FROM t1 GROUP BY id;
+------------------------------------+---------+-------------------+---------------+----------------------------------------------------+
| id | estRows | task | access object | operator info |
+------------------------------------+---------+-------------------+---------------+----------------------------------------------------+
| TableReader_31 | 2.00 | root | | data:ExchangeSender_30 |
| └─ExchangeSender_30 | 2.00 | batchCop[tiflash] | | ExchangeType: PassThrough |
| └─Projection_26 | 2.00 | batchCop[tiflash] | | Column#4 |
| └─HashAgg_27 | 2.00 | batchCop[tiflash] | | group by:test.t1.id, funcs:sum(Column#7)->Column#4 |
| └─ExchangeReceiver_29 | 2.00 | batchCop[tiflash] | | |
| └─ExchangeSender_28 | 2.00 | batchCop[tiflash] | | ExchangeType: HashPartition, Hash Cols: test.t1.id |
| └─HashAgg_9 | 2.00 | batchCop[tiflash] | | group by:test.t1.id, funcs:count(1)->Column#7 |
| └─TableFullScan_25 | 3.00 | batchCop[tiflash] | table:t1 | keep order:false |
+------------------------------------+---------+-------------------+---------------+----------------------------------------------------+
The above execution plan contains two query fragments:
- The first is
[TableFullScan_25, HashAgg_9, ExchangeSender_28]
, which is mainly responsible for the first-stage aggregation. - The second is
[ExchangeReceiver_29, HashAgg_27, Projection_26, ExchangeSender_30]
, which is mainly responsible for the second-stage aggregation.
The operator info
column of the ExchangeSender
operator shows the exchange type information. Currently, there are three exchange types. See the following:
- HashPartition: The
ExchangeSender
operator firstly partitions data according to the Hash values and then distributes data to theExchangeReceiver
operator of upstream MPP tasks. This exchange type is often used for Hash Aggregation and Shuffle Hash Join algorithms. - Broadcast: The
ExchangeSender
operator distributes data to upstream MPP tasks through broadcast. This exchange type is often used for Broadcast Join. - PassThrough: The
ExchangeSender
operator sends data to the only upstream MPP task, which is different from the Broadcast type. This exchange type is often used when returning data to TiDB.
In the example execution plan, the exchange type of the operator ExchangeSender_28
is HashPartition, meaning that it performs the Hash Aggregation algorithm. The exchange type of the operator ExchangeSender_30
is PassThrough, meaning that it is used to return data to TiDB.
MPP is also often applied to join operations. The MPP mode in TiDB supports the following two join algorithms:
- Shuffle Hash Join: Shuffle the data input from the join operation using the HashPartition exchange type. Then, upstream MPP tasks join data within the same partition.
- Broadcast Join: Broadcast data of the small table in the join operation to each node, after which each node joins the data separately.
The following is a typical execution plan for Shuffle Hash Join:
SET tidb_broadcast_join_threshold_count=0;
SET tidb_broadcast_join_threshold_size=0;
EXPLAIN SELECT COUNT(*) FROM t1 a JOIN t1 b ON a.id = b.id;
+----------------------------------------+---------+--------------+---------------+----------------------------------------------------+
| id | estRows | task | access object | operator info |
+----------------------------------------+---------+--------------+---------------+----------------------------------------------------+
| StreamAgg_14 | 1.00 | root | | funcs:count(1)->Column#7 |
| └─TableReader_48 | 9.00 | root | | data:ExchangeSender_47 |
| └─ExchangeSender_47 | 9.00 | cop[tiflash] | | ExchangeType: PassThrough |
| └─HashJoin_44 | 9.00 | cop[tiflash] | | inner join, equal:[eq(test.t1.id, test.t1.id)] |
| ├─ExchangeReceiver_19(Build) | 6.00 | cop[tiflash] | | |
| │ └─ExchangeSender_18 | 6.00 | cop[tiflash] | | ExchangeType: HashPartition, Hash Cols: test.t1.id |
| │ └─Selection_17 | 6.00 | cop[tiflash] | | not(isnull(test.t1.id)) |
| │ └─TableFullScan_16 | 6.00 | cop[tiflash] | table:a | keep order:false |
| └─ExchangeReceiver_23(Probe) | 6.00 | cop[tiflash] | | |
| └─ExchangeSender_22 | 6.00 | cop[tiflash] | | ExchangeType: HashPartition, Hash Cols: test.t1.id |
| └─Selection_21 | 6.00 | cop[tiflash] | | not(isnull(test.t1.id)) |
| └─TableFullScan_20 | 6.00 | cop[tiflash] | table:b | keep order:false |
+----------------------------------------+---------+--------------+---------------+----------------------------------------------------+
12 rows in set (0.00 sec)
In the above execution plan:
- The query fragment
[TableFullScan_20, Selection_21, ExchangeSender_22]
reads data from table b and shuffles data to upstream MPP tasks. - The query fragment
[TableFullScan_16, Selection_17, ExchangeSender_18]
reads data from table a and shuffles data to upstream MPP tasks. - The query fragment
[ExchangeReceiver_19, ExchangeReceiver_23, HashJoin_44, ExchangeSender_47]
joins all data and returns it to TiDB.
A typical execution plan for Broadcast Join is as follows:
EXPLAIN SELECT COUNT(*) FROM t1 a JOIN t1 b ON a.id = b.id;
+----------------------------------------+---------+--------------+---------------+------------------------------------------------+
| id | estRows | task | access object | operator info |
+----------------------------------------+---------+--------------+---------------+------------------------------------------------+
| StreamAgg_15 | 1.00 | root | | funcs:count(1)->Column#7 |
| └─TableReader_47 | 9.00 | root | | data:ExchangeSender_46 |
| └─ExchangeSender_46 | 9.00 | cop[tiflash] | | ExchangeType: PassThrough |
| └─HashJoin_43 | 9.00 | cop[tiflash] | | inner join, equal:[eq(test.t1.id, test.t1.id)] |
| ├─ExchangeReceiver_20(Build) | 6.00 | cop[tiflash] | | |
| │ └─ExchangeSender_19 | 6.00 | cop[tiflash] | | ExchangeType: Broadcast |
| │ └─Selection_18 | 6.00 | cop[tiflash] | | not(isnull(test.t1.id)) |
| │ └─TableFullScan_17 | 6.00 | cop[tiflash] | table:a | keep order:false |
| └─Selection_22(Probe) | 6.00 | cop[tiflash] | | not(isnull(test.t1.id)) |
| └─TableFullScan_21 | 6.00 | cop[tiflash] | table:b | keep order:false |
+----------------------------------------+---------+--------------+---------------+------------------------------------------------+
In the above execution plan:
- The query fragment
[TableFullScan_17, Selection_18, ExchangeSender_19]
reads data from the small table (table a) and broadcasts the data to each node that contains data from the large table (table b). - The query fragment
[TableFullScan_21, Selection_22, ExchangeReceiver_20, HashJoin_43, ExchangeSender_46]
joins all data and returns it to TiDB.
EXPLAIN ANALYZE
statements in the MPP mode
The EXPLAIN ANALYZE
statement is similar to EXPLAIN
, but it also outputs some runtime information.
The following is the output of a simple EXPLAIN ANALYZE
example:
EXPLAIN ANALYZE SELECT COUNT(*) FROM t1 GROUP BY id;
+------------------------------------+---------+---------+-------------------+---------------+---------------------------------------------------------------------------------------------+----------------------------------------------------------------+--------+------+
| id | estRows | actRows | task | access object | execution info | operator info | memory | disk |
+------------------------------------+---------+---------+-------------------+---------------+---------------------------------------------------------------------------------------------+----------------------------------------------------------------+--------+------+
| TableReader_31 | 4.00 | 2 | root | | time:44.5ms, loops:2, cop_task: {num: 1, max: 0s, proc_keys: 0, copr_cache_hit_ratio: 0.00} | data:ExchangeSender_30 | N/A | N/A |
| └─ExchangeSender_30 | 4.00 | 2 | batchCop[tiflash] | | tiflash_task:{time:16.5ms, loops:1, threads:1} | ExchangeType: PassThrough, tasks: [2, 3, 4] | N/A | N/A |
| └─Projection_26 | 4.00 | 2 | batchCop[tiflash] | | tiflash_task:{time:16.5ms, loops:1, threads:1} | Column#4 | N/A | N/A |
| └─HashAgg_27 | 4.00 | 2 | batchCop[tiflash] | | tiflash_task:{time:16.5ms, loops:1, threads:1} | group by:test.t1.id, funcs:sum(Column#7)->Column#4 | N/A | N/A |
| └─ExchangeReceiver_29 | 4.00 | 2 | batchCop[tiflash] | | tiflash_task:{time:14.5ms, loops:1, threads:20} | | N/A | N/A |
| └─ExchangeSender_28 | 4.00 | 0 | batchCop[tiflash] | | tiflash_task:{time:9.49ms, loops:0, threads:0} | ExchangeType: HashPartition, Hash Cols: test.t1.id, tasks: [1] | N/A | N/A |
| └─HashAgg_9 | 4.00 | 0 | batchCop[tiflash] | | tiflash_task:{time:9.49ms, loops:0, threads:0} | group by:test.t1.id, funcs:count(1)->Column#7 | N/A | N/A |
| └─TableFullScan_25 | 6.00 | 0 | batchCop[tiflash] | table:t1 | tiflash_task:{time:9.49ms, loops:0, threads:0} | keep order:false | N/A | N/A |
+------------------------------------+---------+---------+-------------------+---------------+---------------------------------------------------------------------------------------------+----------------------------------------------------------------+--------+------+
Compared to the output of EXPLAIN
, the operator info
column of the operator ExchangeSender
also shows tasks
, which records the id of the MPP task that the query fragment instantiates into. In addition, each MPP operator has a threads
field in the execution info
column, which records the concurrency of operations when TiDB executes this operator. If the cluster consists of multiple nodes, this concurrency is the result of adding up the concurrency of all nodes.