Product Introduction
Cloud Service for etcd is an etcd management solution based on open-source etcd and is tailored to cloud-native service scenarios. It is provided by the TKE team and fully compatible with distributed storage capabilities of open-source etcd. It provides highly stable, observable, Ops-free cloud-native etcd services. Notes:
The beta testing of Cloud Service for etcd has ended on June 8, 2022. The service is now officially available with commercial billing. For pricing details, see Purchase Guide. Use Cases
etcd is a distributed and highly reliable key-value store that can tolerate failures of some nodes in a cluster. It can provide services with more than half of the nodes being alive. It is mainly used for metadata storage, service discovery, distributed election, and other purposes. Features such as release and subscription can be easily implemented based on the Watch mechanism provided by etcd.
Reasons for Using Cloud Service for etcd
Users lack sufficient understanding of etcd, making it difficult to use etcd.
Users lack usage experience in Ops of self-built etcd clusters. They cannot quickly locate issues during the use of them.
Self-built etcd clusters often requires maintaining a monitoring and alarm system, as well as a backup and recovery mechanism, which increases the users' Ops burden.
The TKE team currently maintains tens of thousands of Kubernetes clusters online and uses thousands of etcd clusters for storage at the backend. The team has accumulated a wealth of practical experience in ensuring the stable operation of etcd, which can help users reduce the Ops burden of etcd and focus more on business development.
Product Features
One-click etcd cluster deployment: Support high availability deployment, HTTPS access, and automatic data compression for clusters.
Integration with cloud-native monitoring capabilities: Provide a comprehensive monitoring and alarm mechanism.
Routine Ops management: Support backup and recovery, node scaling, version upgrade, and other features.
Product Strengths
Complete Cluster Protection Mechanism
Secure data access: Support enabling HTTPS two-way authentication for more secure data access. Support restricting access sources through security groups.
Data backup and recovery: Support enabling cluster backup policies to regularly back up data to Tencent Cloud Object Storage (COS), and support recovering faulty clusters from backup data.
High availability deployment: Support deploying cluster nodes to different AZs. If 3 different AZs are used, failures of a single AZ do not affect the normal service of the cluster. Support quick self-healing of faulty nodes to minimize cluster downtime.
Reliable version update mechanism: Carry out comprehensive internal testing and large-scale verification before a new version is released. Conduct fault experiment on Chaotic Fault Generator (CFG) to ensure the stability of a new version.
Powerful Monitoring and Alarm Capabilities
Cloud Service for etcd can be seamlessly integrated with TencentCloud Managed Service for Prometheus (TMP) to provide default performance and availability metrics you want to monitor. You can also aggregate the required monitoring metrics and panels to better monitor the etcd cluster status.
Full-Process Ops Service
Users do not need to worry about the installation, deployment, version updates, and failure handling of Cloud Service for etcd.
Product Pricing
The beta testing of Cloud Service for etcd has ended on June 8, 2022. The service is now officially available with commercial billing. For pricing details, see Purchase Guide. Resources of other products such as Cloud Load Balancer (CLB) and COS will be created automatically during the use of Cloud Service for etcd. For billing details of these products, see the corresponding billing overview, such as CLB Purchase Overview and COS Billing Overview. Cloud Resource | Applicable Scenarios |
CLB instance | Private network access of clusters |
COS instance | Data backup |
Related Services
TMP is a monitoring and alarm solution for cloud-native service scenarios. It fully supports the monitoring capabilities of open-source Prometheus and provides lightweight, stable, and highly available cloud-native Prometheus monitoring services for users.