Monitor multi-cloud container clusters with registered clusters and Managed Service for Prometheus - Application Real-Time Monitoring Service

Learn about the challenges of multi-cloud container cluster observability and how to use Alibaba Cloud Managed Service for Prometheus with registered clusters to centrally manage and monitor container clusters across cloud providers.

Background

As Kubernetes (K8s) has become the de facto industry standard for container orchestration, Prometheus has emerged as the standard for container cluster monitoring after extensive real-world adoption.

Prometheus monitors metrics at the system, application, and business levels. Combined with Grafana, it provides a comprehensive solution for collecting, storing, visualizing, and alerting on container cluster metrics, helping you identify and locate issues to secure your cloud-native applications.

You have two options to operate and maintain container clusters:

Option 1: Build your own monitoring system

You can build a production-ready monitoring system from scratch by using Prometheus and Grafana. This approach requires a significant initial investment in staffing. You must also manage the integration of all components, including metric collection configuration, metric storage, data visualization, effective dashboard configuration, alert configuration, and filtering of invalid alerts. This leads to high ongoing O&M costs.

Option 2: Use a managed service

Alibaba Cloud offers Managed Service for Prometheus, which is available in both subscription and pay-as-you-go billing models. This service reduces the high initial costs of building your own monitoring system and provides ongoing technical and O&M support, which significantly lowers your operational overhead.

Challenges of multi-cloud container cluster observability

As enterprise cloud services grow more diverse and complex, container clusters commonly span different regions and cloud providers. This shifts the operational focus from managing a single cluster to managing multiple clusters in a multi-cloud environment.

You have two options to monitor hybrid, multi-cloud container clusters:

Option 1: Self-hosted Prometheus and Grafana

This approach presents several challenges:

Building a complete monitoring system requires integrating collection, storage, visualization, and alerting components. Ongoing maintenance requires dedicated personnel, increasing O&M costs.
The time-series database (TSDB) in open-source Prometheus stores data locally on a single node. This creates a single point of failure and a risk of data loss.
Open-source Prometheus has collection bottlenecks and runs as a single instance that cannot scale elastically. This may lead to performance issues in data collection during peak hours.

Option 2: Use a managed Prometheus service

This approach also has its challenges:

Cross-cloud compatibility: Different cloud providers offer Prometheus services with varying capabilities and user experiences. This requires you to invest time in learning each platform.
Decentralized management: Using a mix of monitoring products from different cloud providers prevents unified management. This often leads to inefficiency, confusion, and redundant O&M tasks, making it difficult to quickly detect business-critical issues.

Both options share a common problem: scattered monitoring metrics make unified queries, consolidated analysis, and centralized alerting impossible.

Benefits of Managed Service for Prometheus

Alibaba Cloud registered clusters let you centrally manage container clusters hosted outside Alibaba Cloud, addressing decentralized management in multi-cloud environments. Alibaba Cloud Managed Service for Prometheus is a comprehensive container monitoring system that includes metric collection, Grafana visualization, and alerting. It supports both pay-as-you-go and subscription billing models, improving monitoring efficiency while significantly reducing O&M costs compared to self-hosted systems.

Combining Alibaba Cloud registered clusters with Managed Service for Prometheus enables simple, efficient multi-cloud container cluster monitoring. Key benefits include:

Powerful capabilities: Addresses common multi-cloud monitoring issues such as decentralized management, complex setup, low O&M efficiency, and the inability to unify queries and alerts. It provides unified management, configuration, querying, and alerting for all your container clusters, greatly improving monitoring efficiency, reducing staffing costs and repetitive tasks so your O&M team can focus on your business.
Cost-effectiveness: Alibaba Cloud Managed Service for Prometheus offers free collection for basic metrics, which covers the fundamental monitoring needs of your container clusters. For smaller clusters, you can use the pay-as-you-go model to effectively monitor your services while minimizing costs. For more information about the pay-as-you-go billing of Alibaba Cloud Managed Service for Prometheus, see pay-as-you-go. For large-scale clusters, you can choose the subscription model, which can reduce your costs by about 67% compared with the pay-as-you-go model, significantly lowering the expense of monitoring large environments.
Low resource consumption: Managed Service for Prometheus requires you to deploy only a lightweight agent in your cluster. The agent supports automatic elastic scaling. With a resource request of 2 CPU cores and 4 GB of memory, you can collect 6 million metrics. The service discovery module in open-source Prometheus can put significant pressure on a cluster's API server. Alibaba Cloud Managed Service for Prometheus is optimized to relieve this pressure. This approach minimizes resource usage while maximizing the collection of monitoring metrics to protect your business.

Benefit 1: Improved performance

Item	Managed Service for Prometheus	Self-hosted Prometheus
High availability	Supports multiple replicas for collection and storage components with horizontal scaling for high availability.	As a single process that cannot scale horizontally, it has low availability.
Data storage	Based on cloud storage with unlimited capacity.	Limited by the storage capacity of the host.
Data visualization	Includes a built-in Grafana service with out-of-the-box monitoring templates.	Requires you to deploy Grafana separately and requires manual dashboard configuration.
Alert management	Integrates with the ARMS alert center to improve alerting efficiency and precision.	Requires manual integration of the AlertManager plug-in.
Collection performance of a single replica (2 CPU cores, 4 GB memory)	6 million data points	1 million data points
Data query performance (600 million time points)	8 to 10 seconds	180 seconds
Other capabilities	Supports pre-aggregation, downsampling, and Global View.	Not supported.

Benefit 2: Multi-cluster aggregated query

A virtual aggregation instance spans multiple Alibaba Cloud Prometheus instances or self-hosted Prometheus clusters, enabling unified querying, a unified Grafana data source, and centralized alerting for all your Prometheus metrics.

Open-source Prometheus requires configuring multiple data source addresses in Grafana to access scattered data. Aggregated queries let you analyze the global health of your applications from a single, unified perspective.
Instead of deploying a Prometheus server in each region, use Remote Write to send data to Alibaba Cloud Managed Service for Prometheus and use Global View without deploying Thanos components or managing additional dependencies.
The global query feature is built on a distributed, performance-optimized query engine that scales horizontally and vertically on demand, ensuring high performance and stability under large query volumes.
Available out of the box as part of Alibaba Cloud Managed Service for Prometheus, this feature requires no additional components, significantly reducing O&M overhead.

Benefit 3: Lightweight installation

Compared with installing open-source Prometheus, Alibaba Cloud Managed Service for Prometheus is much simpler. You need to install only a lightweight agent in your container cluster. The backend storage is fully managed, which saves resources in your business cluster.

Benefit 4: Integrated Grafana service

Alibaba Cloud Managed Service for Grafana is a maintenance-free, cloud-native data visualization platform that provides a ready-to-use Grafana environment with the following benefits:

Integrates by default with Alibaba Cloud data sources such as Managed Service for Prometheus and Log Service (SLS), and supports third-party or custom data sources for building a unified O&M dashboard.
Provides dedicated instances with a high SLA, ensuring high availability and elasticity for your monitoring system while lowering maintenance costs.
Integrates with Alibaba Cloud account SSO and custom account systems for fine-grained management of data sources and dashboards while ensuring data security.
Solves the following problems:
- Difficulty in data aggregation: Monitoring data from various cloud services is hard to centralize, which increases O&M complexity.
- Complex O&M and monitoring: Core metric charts for various cloud services must be repeatedly configured.
- Difficulty in alert management: Alert rules for different cloud services are scattered and cannot be managed uniformly.
Provides the following capabilities:
- Default integrations: Integrated by default with core Alibaba Cloud services, including Elastic Compute Service (ECS) and databases.
- Unified dashboards: Establishes a unified dashboard system across multiple data sources for more refined visual operations.
- Unified alerting: Build an integrated alerting system to improve alert management efficiency.

Benefit 5: Integrated alerting system

Alibaba Cloud Managed Service for Prometheus is integrated with Alibaba Cloud's unified alerting system by default. This system has the following features:

Global scope
- Use global alert rule templates to configure alerts for events worldwide from a single location.
- Contacts and notification policies are global. Configure them once to apply them everywhere.
More efficient management of integrated events
- Alert management supports one-click integration with common Alibaba Cloud monitoring tools and manual integration with other tools for unified maintenance.
- The event ingestion module is stable and provides 24/7 uninterrupted event processing services.
- It ensures low-latency processing of massive event data.
Timely and accurate notifications to contacts
- Configure notification rules to consolidate events before sending notifications. This reduces notification fatigue for your O&M team.
- Choose different notification methods such as email, SMS, phone call, or DingTalk based on the urgency of the alert.
- Use escalation policies to send multiple reminders for unhandled alerts to ensure timely resolution.
Rapid alert management
- Contacts can handle alerts at any time through DingTalk.
- A universal alert format helps contacts analyze alerts more effectively.
- Multiple contacts can collaborate on handling alerts through DingTalk.
Alert event reprocessing
- Use event processing flows to create simple workflows that reprocess alert events from any source to meet different data handling needs.
- Event management supports deduplication, compression, noise reduction, and silencing for alert events from any source. This helps converge alerts and reduce alert storms.
Alert configuration management
- It provides monitoring templates for common core metrics of container clusters. It also provides an alert template feature that allows you to generate and apply alert templates for quick, large-scale alert configuration.
- It offers a guided, visual alert configuration process with a live preview. You can view which events match the alert conditions in real time and fine-tune your configuration.
Analyze alert data and processing status in real time to improve alert handling efficiency and gain insights into your business operations.

Example: Connecting a multi-cloud container cluster to Managed Service for Prometheus

Prerequisites

Clusters can be connected over public or internal networks. For more information, see FAQ about registered clusters.

Step 1: Create an ACK One registered cluster

Log on to the ACK console. In the left navigation pane, click Clusters.
On the Clusters page, click Create Kubernetes Cluster.
On the ACK One Registered Cluster tab, configure the parameters. For more information about the parameters, see Create an ACK One registered cluster.
In the right-side panel, click Create Cluster. You can view the created registered cluster on the Clusters page. The cluster is in the Pending state.

Step 2: Connect the multi-cloud cluster

This example connects a Tencent Cloud Kubernetes cluster to an ACK One registered cluster for unified management, enabling metric scraping and visualization through Managed Service for Prometheus.

Log on to the ACK console. On the Clusters page, find the registered cluster that you created in Step 1: Create an ACK One registered cluster and click Details in the Operation column.
Click the Connection Information tab. Click Obtain Temporary kubeconfig or Obtain Long-term kubeconfig based on your needs. In the dialog box that appears, select the Internal Access or Public Access tab, and click Copy to copy the cluster credentials.
Log on to the Tencent Cloud container service console. On the Cluster page, click the name of the target cluster. In the upper-right corner, click Create Resource from YAML. Paste the public access credentials that you copied in the previous step into the editor, and then click Finish to install the agent. Then, on the Tencent Cloud Cluster page, check the status of the ack-cluster-agent Deployment. If it is running, the agent was installed successfully.
Log on to the Alibaba Cloud ACK console. On the Clusters page, check the status of the registered cluster created in Step 1: Create an ACK One registered cluster. If the status is Running, the cluster is successfully connected and managed.

Step 3: Install the Prometheus component

For detailed instructions, see Connect an Alibaba Cloud Prometheus instance to a registered cluster.

Step 4: View monitoring data

Managed Service for Prometheus includes default Grafana dashboards for common resources such as Deployments and DaemonSets. To view your monitoring data, perform the following steps:

Log on to the ARMS console. In the left-side navigation pane, choose Managed Service for Prometheus > Instances.
Click the Prometheus instance for the registered cluster that you created in Step 1: Create an ACK One registered cluster.
In the left-side navigation pane, click Dashboards. On the Dashboards page, you can click any dashboard name to view detailed metrics.

Step 5: View alerts

Managed Service for Prometheus monitors core container cluster metrics by default, preventing coverage gaps from manual setup. It includes core metric alert templates that you can enable as needed without writing PromQL queries. To view alerts in Managed Service for Prometheus, perform the following steps:

Log on to the ARMS console. In the left-side navigation pane, choose Managed Service for Prometheus > Instances.
Click the Prometheus instance for the registered cluster that you created in Step 1: Create an ACK One registered cluster.
In the left-side navigation pane, click Alarm Rules. On the Alarm Rules page, you can view alert details.

Activation

Alibaba Cloud registered cluster: For information about how to activate registered clusters, see Activate a registered cluster.
Managed Service for Prometheus: Managed Service for Prometheus offers a subscription billing model that can save you at least 67% compared with the pay-as-you-go model.