Data centers with traditional infrastructure often lack sufficient redundancy and rapid recovery capabilities when hardware failures occur, such as failures of physical servers, networks, or physical disks. A critical data center site-level failure can easily lead to business disruptions, causing significant financial losses for customers. Traditional disaster recovery solutions face challenges such as high construction costs, complex architecture, and lengthy recovery times. These solutions fail to meet the core requirement for fast failback in critical business scenarios—namely, minimal recovery point objective (RPO approaching zero) and rapid, automatic business recovery.
To address the challenges mentioned above, ACOS introduces an active-active cluster solution. This solution integrates the physical nodes of two data centers in the same city with a certain physical distance into a single stretched cluster. This achieves real-time data synchronization across data centers, ensuring no data loss (RPO=0). When a data center site-level failure occurs, ACOS can automatically restore business virtual machines in the failed data center to healthy data centers within minutes, significantly improving system high availability.
