Getting Started with Common Practices

After completing basic operations such as managing applications and containers, you can implement common practices based on this section.

**Table 1** AOM best practices
Best Practice	Description
Preventing ELB Alarm Storms Using AOM Alarm Grouping Rules	This section describes how to set alarm noise reduction. Before sending an alarm notification, AOM processes alarms based on noise reduction rules to prevent alarm storms.
Unified Metric Monitoring Through a Prometheus Instance for Multi-Account Aggregation	This section describes how to centrally monitor metric data of different accounts.
Automatically Connecting to UniAgents Using Custom Images	This section describes how to package images for connecting to UniAgents in the Linux and Windows environments. By using an image, you can purchase an Elastic Compute Server (ECS). A UniAgent will then be automatically installed on it.
Connecting Self-Built Middleware to AOM in the CCE Container Scenario for Metric Monitoring	Prometheus monitoring provides multiple common middleware Exporters. AOM is compatible with native Prometheus. You can install Exporters in the community to connect self-built middleware to AOM in the CCE container scenario.
Interconnecting Self-Built Prometheus of a Third-Party Vendor or IDC with an AOM Prometheus Instance	It is common for cloud users to interconnect self-built Prometheus of a third-party cloud vendor or Internet Data Center (IDC) with an AOM Prometheus instance.
Generating Alarms by Tag	Through Prometheus monitoring and alarm management, alarms can be generated for resources by tag. This section describes how to generate alarms for the CPU usage of a Distributed Cache Service (DCS) instance by tag.
Suggestions on AOM Security Configuration	This section provides guidance for enhancing the overall security of AOM. You can continuously evaluate the security of AOM and combine different security capabilities to enhance overall defense. By doing this, data stored in AOM can be protected from leakage and tampering both at rest and in transit.
Configuring a Recording Rule to Accelerate Metric Query	Based on recording rules, the system calculates the expressions that are frequently used or require heavy calculation in advance and saves the results as a group of new time series. This helps optimize the performance of complex PromQL statements and improve query efficiency. By setting recording rules, you can move the computing process to the write end, reducing resource usage on the query end.
Creating an APM Alarm Rule and Configuring Alarm Notification	AOM allows you to create an APM alarm rule to monitor and manage key metrics in real time. In this way, you will be able to improve application performance and stability.
Configuring Message Templates	AOM allows you to customize notification content by creating message templates. When an alarm notification policy is triggered, the system sends alarm information to specified recipients through emails, SMS, voice calls, HTTP/HTTPS, Lark, WeCom, DingTalk, or WeLink.
Notification Policies	You can create an alarm notification rule and associate it with an SMN topic and a message template. If the log/resource/metric data meets the alarm condition, the system sends an alarm notification based on the associated SMN topic and message template.
Using Prometheus to Monitor MySQL Metrics	MySQL Exporter is designed to collect MySQL database metrics. Core database metrics can be collected through the Exporter for monitoring and alarm sending. This section describes how to use Prometheus to monitor MySQL metrics on AOM.
Ingesting Prometheus Monitoring Data to Local Grafana	After cloud service or CCE cluster metrics are ingested to a Prometheus instance, you can use local Grafana to view the metric data in AOM.
Configuring Service Discovery Using ServiceMonitor	AOM allows you to configure service discovery using ServiceMonitor. With ServiceMonitor, you can specify the namespace to be discovered and select the services to be monitored by matching labels.