During daily O&M of ApsaraDB RDS for MySQL instances, you can use Database Autonomy Service (DAS) to handle instance failures, optimize instance performance, improve efficiency, and reduce O&M costs. DAS is a cloud service that is developed based on machine learning and expert experience. DAS provides self-awareness, self-healing, self-optimization, self-O&M, and self-securing capabilities to simplify database O&M tasks and ensure the stability, security, and efficiency of database services.
Feature description
DAS provides the following features for ApsaraDB RDS for MySQL:
This feature diagnoses your ApsaraDB RDS for MySQL instance and provides visualized diagnostic results.
This feature allows you to specify a time range and view events such as exceptions, optimization events, and auto scaling events within the specified time range.
This feature allows you to view or terminate sessions on your RDS instance. You can also export statistics on the sessions or perform 10-second SQL analysis, SQL throttling, or SQL optimization.
This feature allows you to view the trends and data of performance metrics for your RDS instance in real time.
This feature allows you to view and analyze the latest deadlock in a database.
This feature allows you to view the capacity assessment suggestions, performance capacity, storage usage, and the estimated service time of available storage. This feature can also use machine learning and capacity algorithms to predict storage usage.
This feature allows you to view the storage usage of your RDS instance. For example, you can view the remaining days for which storage is available, the storage usage of individual tables, the tablespace fragments, and the storage exception diagnosis. You can use this feature to identify the storage exceptions at the earliest opportunity to ensure business stability.
This feature allows you to query the static parameter diagnostic results and parameter modification history within the last seven days.
This feature uses the Performance Schema feature to collect statistics on SQL statements, evaluate the database load, identify the root causes of issues, and provide handling suggestions to improve business stability.
This feature allows you to evaluate the loads of your instance and identify the root causes of performance issues. This improves the stability of your instance.
This feature provides various performance metrics and allows you to create custom charts. This feature provides powerful diagnostic capabilities to detect events on your RDS instance at the earliest opportunity and automatically diagnose the events to identify root causes and provide suggestions. You can also specify a time range to manually diagnose the RDS instance to understand the instance performance within the specified time range.
This feature allows you to view the trends of slow SQL queries, the execution status of slow SQL queries, and optimization suggestions for slow SQL queries.
This feature uses the offline data analysis technology to collect statistics on slow SQL queries and analyze the slow SQL queries that are generated on all RDS instances in the previous day at 01:00 every day and automatically adds tags to the slow SQL queries. This way, you can configure settings to automatically classify and prioritize slow SQL queries. This feature also provides governance suggestions and allows you to export data.
This feature is developed based on the full request feature and the SQL Audit feature and provides the following capabilities: Search, SQL Explorer, Security Audit, and Traffic Playback and Stress Test. You can use this feature to obtain information about the SQL statements that are executed. You can use the information to troubleshoot various performance issues, identify the sources of high risks, and verify the instance specifications.
This feature allows you to generate diagnostic reports or view automatically generated reports about instance health, alerts, and slow queries.
This feature inspects and scores all RDS instances at regular intervals. You can specify the RDS instances that you want the feature to inspect and score and select the inspection methods. You can also manually trigger instance scoring. This way, you can understand the performance of your RDS instances.
This feature allows you to specify RDS instances and metrics to monitor and compare the metrics of the RDS instances. You can also configure metric linkage. This way, you can monitor your RDS instances.
This feature sends you notifications when related events are detected. You can specify a notification method based on your business requirements.
This feature automatically analyzes and optimizes your SQL queries and creates indexes. This way, you can quickly resolve issues about slow SQL queries and ensure the optimal performance of your RDS instance.
This feature allows you to specify conditions to trigger automatic SQL throttling. If the specified conditions are met, automatic SQL throttling is triggered to manage the number of requests to your RDS instance and the number of concurrent SQL queries to ensure service availability.
This feature monitors the storage of your RDS instance. When the storage is insufficient, the system automatically expands the storage capacity to ensure business stability.
This feature automatically scales up your RDS instance to handle traffic peaks and ensure business stability. This feature also monitors the CPU utilization of the RDS instance in real time. If the CPU utilization per unit of time decreases, this feature automatically scales down the RDS instance.
Automatic fragment reclamation
This feature is used to optimize the usage of tablespace. After you enable this feature for your RDS instance, the OPTIMIZE TABLE or ALTER TABLE statement is automatically executed on the primary RDS instance to reclaim tablespace fragments.
This feature predicts the values of metrics in the subsequent 24 hours based on the historical data in the most recent 10 days.
This feature scales up the specifications of your RDS instance based on the configured policy and scales down the specifications to the original specifications after the specified scale-up period ends. This feature helps you manage predictable periodic changes in instance loads and reduce costs while meeting business requirements.
Authorization description
To use the features of DAS, you must grant required permissions to RAM users.
Attach a system policy to grant required permissions to RAM users. For more information, see Grant permissions to RAM users.
The following system policies are provided for DAS:
AliyunHDMFullAccess: grants the RAM user full permissions to manage DAS.
AliyunHDMReadOnlyAccess: grants the RAM user read-only permissions on DAS.
AliyunHDMReadOnlyWithSQLLogArchiveAccess: grants the RAM user read-only permissions on DAS and the permissions to use the search capability of the SQL Explorer and Audit feature. The search capability allows you to export data.
Create a custom policy to grant required permissions to RAM users. For more information, see Create a custom policy.
The following example describes how to grant RAM users read-only permissions on DAS.
{ "Action": [ "hdm:Get*", "hdm:Describe*", "hdm:Query*", "hdm:Support*" ], "Effect": "Allow", "Resource": "acs:rds:*:*:*" }