Realtime Compute for Apache Flink allows you to use CloudMonitor or Managed Service for Prometheus of Application Real-Time Monitoring Service (ARMS) to implement monitoring and alerting features for deployments. CloudMonitor is free of charge. This topic describes the differences between CloudMonitor and Managed Service for Prometheus of ARMS that you use in Realtime Compute for Apache Flink. You can use a service based on your business requirements.
Feature comparison
Category | Item | Managed Service for Prometheus of ARMS | CloudMonitor |
Service availability | N/A | Ensures service availability of at least 99.90% for the monitoring and alerting features. For more information, see Managed Service for Prometheus SLA. | Ensures service availability of at least 99.9% for the alerting feature but does not ensure service availability for the monitoring feature. For more information, see CloudMonitor SLA. |
Cost | Monitoring and alerting fees | You are charged for metric collection and alerting on a pay-as-you-go basis. | No fees for subscriptions, O&M operations, and metric and alert management are charged. For alert notification, you are charged only for alert text messages and alert phone calls based on the number of messages and calls at a low price. |
Data retention period | Provides a free quota of 50 GB per month and retains data for 90 days by default. For more information, see Billing overview. | No storage fees are charged and retains data for 30 days by default. | |
Metrics | Display of metrics | Displays metrics in the monitoring charts in the development console of Realtime Compute for Apache Flink. Allows you to query data by using the PromQL syntax and view data on Grafana dashboards in the ARMS console. | Pre-aggregates metrics. Metrics are displayed in the monitoring charts in the development console of Realtime Compute for Apache Flink and the CloudMonitor console. For more information about the limits of CloudMonitor compared with ARMS, see the "What are the limits of the alerting feature of CloudMonitor compared with ARMS?" section of the FAQ about the monitoring and alerting features and logs topic. |
Number of metrics | For more information about the number of metrics that you can configure for Realtime Compute for Apache Flink deployments in the CloudMonitor console, see Metrics. | ||
Alert management | Configuration methods |
| Allows you to redirect to the CloudMonitor console from the development console of Realtime Compute for Apache Flink for configuration or subscription. |
Number of metrics for which you can configure alert rules |
| Allows you to configure alert rules for a maximum of 20 metrics and subscribe to event-triggered alerts, including alerts for deployment failure events, ECS failure handling events, and ECS proactive O&M events. | |
Alerts for a single metric or multiple metrics |
|
| |
Alert rules |
| Allows you to create an alert rule in the CloudMonitor console. | |
Alert templates | Allows you to create an alert template in the CloudMonitor console. | ||
Event-triggered alerts | Does not support event-triggered alerts other than alerts for deployment failure events. |
| |
Alert notification | Alert notification methods |
| Alert notifications can be sent to contacts by using phone calls, text messages, emails, DingTalk, WeCom, Lark, and webhook. CloudMonitor supports scheduling notifications. For more information, see Alert contacts. |
Alert notification policies |
| Supports dynamic threshold-triggered alerting, merging of alert notifications, and blacklist policies for alert notifications. For more information, see Dynamic threshold-triggered alert rules, Merge alert notifications, and Manage blacklist policies. | |
Alert callback | Supported. | Supported. | |
Initiative alerting | Not supported. | Supported. | |
API operation | Monitoring and alerting API operations | Note For more information about the metrics, see Flink metrics. |