Comparison Between AOM 2.0 and Cloud Eye

This section compares the cloud service monitoring functions of AOM 2.0 and Cloud Eye.

AOM metric data comes from Cloud Eye. AOM's metric data is in Prometheus format while Cloud Eye's metric data is in a custom format. Table 1 compares the cloud service monitoring functions of AOM and Cloud Eye.

**Table 1** Comparing the cloud service monitoring functions of AOM and Cloud Eye
Function	Cloud Eye	AOM 2.0
Unified monitoring across accounts	Not supported.	Supported.
Customization of data storage duration	Not supported (default: 3 months).	Supported (up to 367 days).
Data export	Aggregated data of the last three months can be exported. Raw data of the last 48 hours can be exported.	Dashboards and APIs can be exported. Data can be written to Kafka.
Aggregate query	Only simple query is supported.	Multi-instance aggregation query is supported. For example, aggregation by tag or resource group.
PromQL syntax	Not supported.	Supported when you use alarm rules and dashboards, and browse metrics.
Dashboards	Single-instance dashboards are supported for standard cloud products.	Various preset templates are provided.
Graph types supported by dashboards	2	8+
Monitoring views supported by a dashboard	50	100+
Alarm rules that can be created	Max.: 1000.	Default: 3000+. More than 10,000 rules can be supported.
Alarm rules that can be added to an alarm template	Max.: 50.	More than 20 cloud services can be added, and more than 100 alarm rules can be added for each cloud service.
Time that the alarm history can be kept	7 days.	1 year.
Objects that can be selected for single alarm rule creation	5000	Not limited. You can select all resources, and implement regular expression or exact match.
Alarm aggregation	Not supported.	Alarm aggregation based on PromQL syntax is supported. For example, implement alarm aggregation based on the total Content Delivery Network (CDN) bandwidth, total Elastic IP (EIP) bandwidth, or Object Storage Service (OBS) storage condition.
Connecting to the on-premises Grafana	Not supported.	Prometheus data sources can be directly connected to on-premises Grafana.
Interconnecting with on-premises self-built Prometheus	Not supported.	Data can be directly written to self-built Prometheus.
Business monitoring	Not supported.	Business monitoring based on Prometheus, LTS logs, and custom channels is supported.
Application monitoring	Not supported.	JVM, Spring, Nginx, Tengine, and Tomcat application monitoring is supported.
On-premises IDC monitoring	Not supported.	Prometheus Exporter-based on-premises hardware, storage, and network monitoring is supported.
On-premises middleware monitoring	Not supported.	On-premises middleware such as MongoDB, Redis, and RocketMQ can be monitored.