Help Center/ Well-Architected Framework/ Well-Architected Framework and Practices/ Resilience Pillar/ Cloud Service Reliability/ ECS/ Reliability Functions
Updated on 2025-05-22 GMT+08:00
View PDF
Share

Reliability Functions

Data Backup and Restoration

You can use Cloud Backup and Recovery (CBR) to back up and restore ECSs. CBR works based on the consistent snapshots for EVS disks.

For details, see CBR Overview.

Automatic Fault Recovery

You can enable automatic recovery for ECSs. This will help ensure high availability and make it easier for them to be dynamically migrated. If a physical host accommodating ECSs breaks down, the ECSs will automatically be migrated to a functional physical host to minimize the impact on your services. During the process, the ECSs will restart.

For details, see Can ECSs Automatically Recover After the Physical Host Accommodating the ECSs Becomes Faulty?

When the system detects that a physical host accommodating ECSs is sub-healthy, the system automatically migrates the ECSs to another available physical host. During the migration, there is a brief service interruption of less than 1 second, and the performance is impacted.

Anti-Affinity

An ECS group allows you to create ECSs on different physical hosts to improve service reliability.

For details, see Managing ECS Groups.

Cluster HA

You can use a shared EVS disk to deploy an intra-AZ cluster or a mission-critical application with HA enabled. One shared EVS disk can be attached to up to 16 ECSs. You also need to deploy a shared file system or a cluster management system, such as Windows MSCS, Veritas VCS, or CFS.

For higher reliability, you are advised to attach a shared EVS disk to ECSs in the same anti-affinity ECS group.

For details, see Shared EVS Disks and Usage Instructions.

Load Balancing

You can use ELB to enable load balancing across multiple ECSs.

Health Checks

You can use ELB to perform health checks on ECSs.

For details, see Configuring a Health Check.

Auto Scaling

With Auto Scaling, you can scale ECSs to balance loads across AZs.

Cross-AZ DR

You can use ELB to enable failovers across AZs.

Monitoring

Using Cloud Eye, you can monitor ECS resources such as vCPUs, memory, disks, and networks.

For details, see Monitoring ECSs.

Parent topic: ECS