RoseMirrorHA

The arbitration service is responsible for making decisions and coordinating the active and standby roles in the cluster, ensuring maximum continuity of business system.

The arbitration service can balance the requirements of business continuity and data consistency, enhancing the fault response and self-resuming capabilities of the cluster system, thus improving its availability. The arbitration node supports cooperation with multiple cluster groups simultaneously, reducing the deployment complexity of the arbitration node.

It supports dual-engine work modes:  byte-level and block-level. User can simultaneously configure byte-level and block-level data replication mode in a solution, which is suitable for more complex data scenario. 

Based on the block-level real-time replication of logic volume, it does not care about the file type or file-level IO. Instead, it only replicates the changed IO block on production volume, which is suitable for the data scenario with large number of small-size files or big data volume. 

The application service is online, and it supports real-time and continuous data mirroring; any IO change of data source will be replicated to target server, to ensure the data consistency between source and target at any time point.  

According to user demand, it supports user-defined data file or directory replication. And user can dynamically add or reduce replication scope.

It supports online upgrade to disaster recovery solution, which can promote data security protection level and promote cluster’s data security protection level and further satisfy the requirement of Cybersecurity Classified Protection Standards 2.0.

The server HW configuration requirement for data mirroring cluster is not high, and it can be built by two servers. User can apply servers in different specifications or configurations to save investment. 

Can widely support the operation environment for both physical and virtual servers. The deployment modes are quite flexible, like P-P，V-V，P-V and V-P.

It monitors the running status of application service in real time. In case of any abnormity in application service, it will auto failover it to standby server.

Conduct in-depth monitoring on the usage of CPU/memory resource, and support intelligent alarm and failover. 

Continuously monitor the network resources. If the business NIC or network resources fail, it will auto failover related resources to standby server.

The user interface is based on browser, can provide more convenient management and monitoring for user.

Apply friendly wizard to guide configuration. The installation and configuration are simplified. 

It supports cross-platform remote management. For example, user can manage the active-standby cluster on Linux platform from a Windows system.

Under Window/Linux environment, it can auto generate snapshot data. User can recover history snapshot to prevent data loss caused by man-made error, application logic error or ransomware infection. 

Support the latest driver architecture of Microsoft, and can be better compatible with anti-virus software, backup software, encryption software and etc.

Support localized CPU architectures, such as Loongson, Phytium, Kunpeng, Hygon and etc.