TDSQL-C for MySQL (TDSQL-C for MySQL) is a self-developed new-generation cloud-native relational database by Tencent Cloud. It integrates the advantages of traditional databases, cloud computing, and new hardware technologies, is 100% compatible with MySQL, and provides users with flexible elasticity, high performance, high availability, high reliability, and secure database services. It achieves high throughput of over one million QPS, PB-level massive distributed intelligent storage, and Serverless second-level scaling, helping enterprises accelerate their digital transformation.
TDSQL-C for MySQL provides a comprehensive solution for database Ops including backup, recovery, monitoring, rapid scaling, data transmission, and so on, simplifying your IT Ops work and allowing you to focus more on business development.
TDSQL-C for MySQL, after continuous testing and optimization by a professional team, provides various MySQL Enterprise Edition features. Its engine kernel has been extensively optimized to deliver flexible and efficient capabilities of processing transactions, advanced comprehensive compliance and security protection, and ultra-large instance capacity, enabling superior and robust performance.
This section primarily introduces the aspects of performance testing for TDSQL-C for MySQL, including test environment, testing tools, testing methods, test results, and so on. It targets two dataset characteristics—fully cached and large dataset—and conducts performance testing in read-only, mixed read/write, and write-only scenarios, thereby showcasing the overall performance of TDSQL-C for MySQL.
Note:
This section refers to regular clusters as those created via the purchase page. For the creation method, see Create a Cluster. This section refers to the compilation-optimized high-performance version as an optimized kernel version. Currently, access to this version must be requested through the ticket system. For details about this version, see Compilation-Optimized High-Performance Version. Performance Testing Section Overview
|
Testing Elements | | Introducing the environment and information about the test object used in performance testing. |
| | Introducing the testing tools used in performance testing and how to install them on CVM instances. |
| | Introducing the testing methods for performance testing, including running commands and parameter explanations. |
| | Introducing the test metrics for performance testing. |
Test Results (Conventional Cluster) | | Introducing the performance test results of conventional clusters under read-only, mixed read/write, and write-only scenarios with fully cached dataset characteristics. |
| | Introducing the performance test results of conventional clusters under read-only, mixed read/write, and write-only scenarios with large dataset characteristics. |
Test Results (Compilation-Optimized High-Performance Edition) | | Introducing the performance test results of the compilation-optimized high-performance edition under read-only, mixed read/write, and write-only scenarios with fully cached dataset characteristics. |
| | Introducing the performance test results of the compilation-optimized high-performance edition under read-only, mixed read/write, and write-only scenarios with large dataset characteristics. |
Test Scenarios and Read Types
This performance test targets the testing scenarios for fully cached and large datasets, and their corresponding read types are shown in the table below.
Note:
In the table, range select and point select are defined as follows:
range select: Range test, which indicates the number of queries for tests for range selection in a single transaction.
point select: Point test, which indicates the number of queries for tests for point selection in a single transaction.
|
Fully-Cached large dataset | Read-Only | enable binlog | range select |
|
|
| point select |
| Mixed Read/Write | enable binlog | range select |
|
|
| point select |
| Write-Only | enable binlog | - |
Test Results
Test results for conventional clusters are as follows:
Test results for the high-performance version with compilation optimization are as follows: