With the continuous popularization and application of database management system, enterprises require that important data of database will not cause data loss or data being unavailable because of the occurrence of various accidents. Correspondingly, the backup and recovery mechanism of database data shows its importance more and more. It is not difficult to imagine, finance, taxation, construction engineering all walks of life if there is no backup of the database, once the database corruption, data is not used, the normal business work of enterprises to bring about the consequences of disaster!
Of course, there are many ways to back up data now, such as backing up hardware devices that contain data, mirroring hardware devices, and backing up databases separately in Sybase systems, and so on. But it can be used to build a database backup product that is economical, reliable, high-performance and avoids natural disasters, when the Sybase replication server is the most advanced. By leveraging a secure remote update model, remote nodes can update/replicate data in real time, enabling Sybase to replicate replicated data across different platforms.
The following is a schematic diagram of using Sybase replication servers for database backups:
Advantages of replicating servers:
Real-time replication. When the database in the main node as is updated, the transaction log Transport Manager in the replication server sends the update operation to the node in real time, completes the database update from the node, and maintains the synchronization of the database data on the master-slave node as.
High availability. Once the active database fails for a variety of reasons, you can switch to the backup database, make it the active database, and then process the failed database. In addition to a little interruption, the user's operation of the database is hardly affected.
High fault tolerant ability. The replication server still works after the network fails. Through a set of advanced, intelligent storage and forwarding mechanism to ensure the availability of the system, and in the network connectivity, the system can automatically from the last time the breakpoint sent to continue to save the user's network resources, reduce the delivery time. This intelligent mechanism is ideal for copying large text data.
The backup of the database can be in the LAN scope, but also through the Internet Internet in the complete implementation between the offsite, which is particularly important for the recovery of the large catastrophic database. Because of this, replication servers are widely concerned and applied, the United States ' 911 ' incident, because the Sybase replication server installation, so that a large number of important system data intact, to avoid the resulting catastrophic data loss.
Economical and practical. In the actual use of database backups, replication services can be implemented on 3 server machines (Replication Service Manager RSM, primary node server active, standby node server, respectively), and replication services can be implemented on 2 server machines (RSM and active on 1 servers). You can also implement replication services for different as on 1 server machines (RSM, Active, standby on 1 servers), and users can choose to configure the machine structure flexibly. There are no special requirements for computer machine configuration, such as for NT platforms:
Operating system requirements |
Project |
Requirements |
Cpu |
Pentium processor |
Ram |
Minimum 32MB RAM Replication Service Manager RSM client minimum 8MB |
Disk space |
At least 150MB of available hard disk space |
Operating system |
Windows NT4.0 Service Pack 3, Windows2000 Service Pack 1 Windows 98 can only be used for RSM Client |
Other Hardware |
Recommend at least 32-bit NIC, Nic is required |
For other Unix operating systems, please call the Times Chaoyang database inquiries!
Replication servers also support replication between heterogeneous databases through replication agents, such as Sybase and Oracle, DB2, MVS, Lotus Notes, ODBC, and so on.
Replication systems do not have an impact on the performance of existing systems. Each part of the Sybase replication server does not use the features and functions of the database system, where the components such as Log Transfer Manager work completely independently of the source database. The Sybase Replication Service does not use database triggers and rules, and therefore does not increase the burden on the source database. In addition, the Sybase replication system does not limit the design and planning of local applications for each remote node, nor does it interfere with the optimization of local data access, the design of database indexes, or the distribution of data on disk, and therefore does not increase the complexity of existing systems.