tencent cloud

TDSQL-C for MySQL

DocumentationTDSQL-C for MySQLOperation GuideParallel QuerySetting Parallel QueryParallel Query Metrics

Parallel Query Metrics

Download
Focus Mode
Font Size
Last updated: 2024-07-31 10:03:33
This document describes the monitoring metrics offered by TDSQL-C for MySQL to help you monitor the operations related to parallel database query in real time.

Prerequisites

Database version: TDSQL-C for MySQL 8.0 on kernel version 3.1.8 or later.

Parallel Query of Monitoring Metrics

Metric
Parameter
Unit
Description
Data Aggregation Method
Parallelly Queried Threads
txsql_parallel_threads_currently_used
-
The number of threads currently used for parallel queries.
MAX
Failed Parallel Queries
txsql_parallel_stmt_error
-
The number of parallel query statements that report errors.
SUM
Executed Parallel Queries
txsql_parallel_stmt_executed
-
The number of executed parallel query statements.
SUM
Parallel-to-Serial Queries
txsql_parallel_stmt_fallback
-
The number of parallel-to-serial query statements.
SUM

Practical Tutorial for Monitoring Metrics

The following lists some special issues for which you should check the metrics and their solutions. Example 1: The value of Parallelly Queried Threads is always equal to the value set by txsql_max_parallel_worker_threads. Solution: This example indicates that the threads are always fully loaded. We recommend you increase the value of the txsql_max_parallel_worker_threads parameter when the CPU utilization is low.
Example 2: The value of Failed Parallel Queries increases. Solution: Check the CPU utilization and memory utilization. If the load becomes significantly higher, we recommend you reduce the value of txsql_parallel_degree to 1/4 of the number of CPU cores.
Example 3: The value of Parallel-to-Serial Queries increases. Solution: This example indicates that the currently executed SQL statement does not meet the conditions for parallel query. We recommend you increase the total number of threads (txsql_max_parallel_worker_threads) or the maximum memory (txsql_optimizer_context_max_mem_size) of the parallel query plan environment that a single statement can apply for, so as to ensure that parallel query can be used for SQL statements.
Previous Topic: Viewing Parallel QueryNext Topic: Performance Test Report

Help and Support

Was this page helpful?

Help us improve! Rate your documentation experience in 5 mins.

Feedback