Multi-availability Zone Disaster Recovery
Cross-availability zone deployment is supported with the master machine and slave machine located at different availability zones in one city where real-time data replication is performed via Tencent's direct connect. The master machine is in local availability zones and the slave machine is in remote ones. When the data are accessed externally, local instances are accessed first. If they are down or inaccessible, remote slave machines will be accessed. The cross-availability zone deployment provides MySQL with multi-availability zone disaster recovery. The switching process between the master machine and slave machine is transparent to the user. This eliminates the operational risks of single IDC deployment, and the IDC downtime is completely transparent to the services.
Remote Disaster Recovery Instances
Cross-region remote disaster recovery instances are supported to help you set up remote data synchronization for disaster recovery via the private network. In the event that the IDC where the master MySQL instances reside is down or unavailable due to other reasons, the remote disaster recovery instances enable you to quickly recover data access, allowing you to fast set up a business system remotely and redirect access traffic to remote instances. In this way, you can recover business operation, reduce business losses due to downtime and ensure high database availability in case of disasters. During the data synchronization from master instances to the disaster recovery instances, data write performance will be limited. However, you can upgrade the disaster recovery instances to master instances in the Web console and disconnect them from the data synchronization to resume full performance. The switching process takes only a few seconds, minimizing the impact of business interruption in the disaster.
Financial Solution for Three IDCs in Two Regions
MySQL offers deployment architectures for two IDCs in one city and three IDCs in two regions. When a failure occurs, you can restore the normal operation of your databases in minutes. The solution for three IDCs in two regions can be achieved with just a few clicks on the configuration page: Select MySQL cross-availability zone deployment and strong sync data replication to achieve the solution for two IDCs in one city with strong data consistency; Add a remote disaster recovery node to the cluster to achieve the architecture for three IDCs in two regions.
Completed Hot Backup and Cold Backup
Physical machines of identical architectures are used to provide real-time hot backup. Upon master failure, the slave will be switched to in seconds, improving database availability and ensuring business continuity. Based on cold backup and binlog files, lossless recovery within 3 days and cold backup data dump within 5 days are provided, making secure data recovery available.
Automatic Disaster Recovery
Automatic downtime detection and failover are supported. During master/slave switch, VIP blocking switch process is used, so the switch and failover processes are totally transparent to you. You don't need to worry about the issues caused by failure and downtime of physical machines where the data resides, and you can finish the hot backup switch process without changing the application layer. The disaster recovery switch process is imperceptive to the business.