Deployment and O&M-PolarDB(PolarDB)-阿里云帮助中心

This topic describes the deployment models and O&M capabilities of PolarDB-X, including automated O&M, monitoring, diagnostics and optimization, and backup and recovery.

Deployment models

PolarDB-X is a commercial cloud product on Alibaba Cloud that provides comprehensive support. This includes multiple deployment models, a complete API, and a suite of integrated ecosystem tools.

The Database Platform as a Service (dbPaaS) architecture of PolarDB-X provides unified O&M for database instances. In addition to the built-in user console, you can integrate your own management systems using the API.

The dbPaaS architecture of PolarDB-X is built on Kubernetes and supports multiple deployment models for both public cloud and hybrid cloud environments. Deploying on a public cloud (or a large-scale industry cloud) offers the most flexibility. For hybrid cloud deployments, you can choose between Apsara Stack Enterprise and the more lightweight DBStack. DBStack can run in a production environment with as few as three servers. 456789

Automated O&M

The PolarDB-X dbPaaS console integrates features for creating and releasing instances, backup and recovery, configuration management, monitoring and alerting, and diagnostics and optimization. You can use the console to manage databases, users, permissions, and parameters, and to perform day-to-day operational tasks like data backups and setting up alerts. To simplify integrating PolarDB-X with other systems, the PolarDB-X dbPaaS platform also provides a comprehensive API for instance management, database management, querying monitoring data, and parameter management. For a complete list of API operations, see API overview.

Monitoring, diagnostics, and optimization

Resource Monitoring

In the PolarDB-X console, you can view monitoring information for compute resources, storage resources, and databases. You can also add alerting rules based on these metrics.

Compute resource monitoring provides metrics for compute nodes, including CPU utilization, memory usage, inbound and outbound network traffic, QPS, query latency, number of frontend connections, and number of active threads.
Storage resource monitoring provides metrics for storage nodes, including CPU utilization, memory usage, disk space usage, inbound and outbound network traffic, number of connections, number of active threads, IOPS, QPS, TPS, buffer pool usage, and the number of temporary tables created. On the Storage Resource Monitoring page, you can switch between the Instance, Cluster, and Node tabs to view monitoring data from different dimensions. You can also click Set Alerting Rules to configure alerts.

SQL analysis and optimization

PolarDB-X supports real-time querying, visual analysis, and alert configuration for SQL logs. You can enable this feature in the console to perform real-time SQL log auditing and analysis. You can also create custom monitoring and alerting rules based on specific metrics to receive notifications through various channels when anomalies occur.

You can view and terminate current instance sessions in the console. The Session Statistics page displays aggregated statistics, such as by client and by namespace. The 10-second SQL analysis feature samples sessions for 10 seconds (once per second) and then provides a summary of the results. This analysis helps you easily identify the most frequently executed queries and detect slow queries.

The slow query log feature helps you quickly identify critical SQL queries that affect database performance. It provides an interactive page where you can view slow query templates, detailed information about each slow query, and key metrics such as execution count, duration, and the number of rows returned. The slow query diagnostics and optimization feature also suggests optimizations, such as creating a local index or a global index, based on the SQL execution plan.

To prevent a few SQL queries from consuming excessive resources and affecting overall system performance, the SQL query throttling feature limits the concurrency of specific queries. You can create throttling matching rules in the console to set limits such as maximum concurrency and maximum waiting queue length.

Backup and recovery

Consistent backup and recovery

Consistent backup and recovery provides instance-level point-in-time recovery (PITR) with second-level precision, using periodic full snapshot backups and incremental binlogs. 一致性备份恢复

Table Recycle Bin

When the table recycle bin feature is enabled, tables that are dropped or truncated are temporarily moved to the recycle bin. If you need to recover a table within 2 hours, you can find it on the Table Recycle Bin page in the console. 表回收站

SQL Flashback

The SQL flashback feature lets you undo operations run by mistake. For example, if you accidentally delete data with a DELETE statement, SQL flashback can generate the corresponding INSERT statement to restore the data. SQL 闪回

Data import and export

Binlog service

PolarDB-X is compatible with the MySQL binlog replication protocol. You can treat a PolarDB-X instance as a standard MySQL node, and use other MySQL nodes as the synchronization source or destination for PolarDB-X.

The Change Data Capture (CDC) component in PolarDB-X provides change logs that are compatible with the MySQL binlog format. It abstracts away distributed features such as instance scaling, distributed transactions, and global indexes. You can also use open-source tools like Canal to synchronize data written to PolarDB-X to other storage systems.

MySQL ecosystem tools

Compatibility with MySQL and its ecosystem is a core design goal of PolarDB-X. Its communication protocol and SQL syntax are compatible with MySQL, and tools such as mysqldump and LOAD DATA INFILE work out of the box.

PolarDB-X is also compatible with Alibaba Cloud's database ecosystem tools, including Data Management (DMS), Data Transmission Service (DTS), Database Backup (DBS), Database Autonomy Service (DAS), and DataWorks.