Human society integration system of Chun’an covers 450K people all over the county, and 10K . Therefore, this important system that government developed for people's livelihood cannot suffer any disaster. Considering the need of disaster recovery and the unavailability of the production system in Hangzhou caused by severe breakdown,it’s necessary to build a local backup system in Chun’an to take over key business and minimize the time of breakdown. Through the build of remote backup system for Human society integration system of Chun’an, the risk of key business breakdown, data disaster and data lost are minimized.
The core application of client’s current information system is a Oracle RAC environment built for 4 cloud host servers, here’s the structure block of it:

Production system uses a Oracle RAC environment built for 4 cloud host servers to ensure the steady working of the service.
The application of production system are running on Huatong cloud to ensure the stability of their APP, and is connecting with the back-end database through middleware.
Now the plan is to build a local disaster recovery center to achieve the remote disaster recovery of database and APP.
Achieve the remote disaster recovery of database application through byte level replication.
Summary of entire structure
The priority of this project is to choose a product that can provide excellent, high reliability design, complete storage backup products to control the predictable breakdown situation(update of OS, backup and update of software, etc) and unpredictable breakdown situation(breakdown of software and hardware, operate miss and natural disaster, etc), and a solution for enterprise business continuity, as well as strict quality control and a device supplier that provides excellent 7×24h after-sale service and its solution.
Aiming at the current situation of client, we suggest an application level disaster recovery solution, which can achieve flexible real-time replication and high availability in heterogeneous platform of key database system as well as application servers, in order to achieve the business continuity.
The local disaster backup server of the Oracle database system run in production server room is a x86 host server, and achieve the real-time replication and high availability of the database through Oracle database synchronization software.
Disaster recovery software are applied to achieve the real-time replication and high availability of remote data of the APP system.

The illustration of disaster recovery realization
The disaster recovery realization process of Oracle RAC database:
I2soft i2Active are applied in core Oracle RAC database system, and is deployed separately in database system in Huashu server room in Hangzhou and database system in Human Society server room in Chun’an. The trade statement and data are acquired through online analysis of Oracle redo log, in order to complete schema or table level data replication, as well as full database level replication and replicate data to Oracle database of Chun’an Human Society disaster recovery at the same time.
I2soft i2Active database synchronization software can read the data block online rather than wait for Oracle redo log filed to handle. I2soft i2Active won’t transfer all the Oracle redo log to target server, it don’t deal with information except DML/DDL operations involving appointed replication target (datasheet). Since there is only transmission of trade instructions during the entire process, the amount of data transferred are small, which ensure data replication of low delay in narrow bandwidth environment. In addition, i2Active communicate through standard IP network, so the distance of Oracle database in target server has no limit.
Oracle redo log based on raw device or file system(include ocfs system) can be normally read, including redo log saved in ASM.
With i2soft i2Active, the local database of Chun’an Human Society disaster recovery system stays open and can provide query service. If planned or unplanned breakdown occurs, the work can be swithced to disaster recovery database manually and swiftly after Chun’an Human Society gives a final confirm. Compared to other physical replication methods based on disk or file system, not only does our system save the long waiting time of database recovery and start-up, but also ensures a 100% success of switch.
The disaster recovery realization process of HIS and medical insurance deal APP
High availability of Huatong cloud platform and 2 local physical server is achieved for HIS and medical insurance deal APP through i2Availibility, and the switching method is set to manual switch mode(i2Availibility supports automatic switch as well). Once the APP on government cloud crushes and reaches a certain switching threshold value, the APP can be switched to disaster recovery APP server manually after Chun’an Human Society gives a final confirm. i2Availibility can achieve swift switch on application level and reduce the time of service’s pause in order to keep high availability of service APP.