Today, when information systems are generally the lifeblood of enterprise operations, the establishment of a remote disaster recovery mechanism to ensure business continuity has been generally recognized by enterprise managers. However, there are many types of remote disaster recovery technologies, and most of the architectures are very complicated. Not only the investment in hardware and software equipment and time costs are huge, but the subsequent maintenance costs are also scary. Therefore, although there are many interested builders, most of them are hesitant.
Remote disaster recovery based on backup tapes is currently widely used by enterprises. Managers only need to regularly back up data to offline tapes, and then transport them to remote storage through human labor. Once a disaster occurs, the IT Repair the facility, and then restore the system and data from the tape. Although this method of remote disaster recovery has the lowest cost, the operational losses caused by the long recovery time (usually requiring days or even weeks), as well as the data loss caused by the non-real-time nature of backup / recovery time, have obviously exceeded The range that can be tolerated by enterprises today. The ideal remote disaster recovery architecture should be based on an unlimited wide-area network, and use remote data replication (Remote ReplicaTIon) technology for online (on-line) remote disaster recovery, so as to meet the requirements of enterprise RTO and RPO.
Through the connection of the network, it is not difficult to copy the local data to remote storage, but under a complex IT architecture, it is not easy to achieve success. After several years of technological development, online remote disaster recovery has produced the following two types: host-based (Host based) and storage system (Storage based).
Host-type remote disaster recovery: very limited
Host-type remote disaster recovery, in simple terms, uses TCP / IP through data replication software (such as Veritas Volume Replicator) installed on the server, or data replication / disaster recovery tools (such as Oracle Flash Recovery Area) provided by applications The network connects to the remote backup server and performs data replication in an asynchronous manner.
The advantage of host-type remote disaster recovery is that in an environment with fewer servers, the cost is lower, users do not need to replace too many existing system architectures, and do not have to worry about the compatibility of the back-end storage system, only need to pay Software licensing fees and hardware equipment costs at the disaster recovery end are sufficient. However, in an environment with a large number of servers, the complexity of management will increase, and the overall input cost will also increase. Another disadvantage of it is that the software is installed on the application host, and it will consume the running resources of the host when running. If the hardware level is not high, it may affect the application.
Storage system type remote disaster recovery: expensive
Storage system type remote disaster recovery, as the name suggests, is based on the storage system (fiber disk array, NAS) mode. Through the built-in firmware (firmware) or operating system of the storage system, through the IP network or DWDM, Fibre Channel and other transmission interface links, the data is copied to the remote side in a synchronous or asynchronous manner. Well-known remote disaster recovery solutions for storage systems include SRDF, TrueCopy, PPRC, etc.
Compared with the host-type remote disaster recovery, the advantage of the storage system-based remote disaster recovery is that it separates the data from the operation, and has little impact on the operating resources of the host system. In addition, because most of the operating mechanisms use mirrors to copy data and use cache memory to speed up I / O access, the data difference between the two ends is relatively small, and the storage system itself has certain fault tolerance and certain Operating performance and reliability.
The biggest limitation of storage system type remote disaster recovery is its expensive construction cost. Since users must configure two sets of the same storage system at the local end and the disaster recovery end, not only is the procurement cost high, but also subject to a single equipment manufacturer, the future scalability is bound to lack flexibility. In addition, if the Fibre Channel storage system is to build remote disaster recovery, one FC-to-IP adapter must be installed on the local end and the disaster recovery end, and the hardware cost will exceed US $ 50,000, plus the network bandwidth cost. The overall cost input will be staggering. If the enterprise does not undergo a careful evaluation before installation, building a storage system type remote disaster recovery is likely to cause a black hole in IT expenditure and increase the financial burden. In addition, the storage system type disaster recovery method has a great defect for the consistent disaster recovery of the database, and it is not applicable to the multi-point to one-point disaster recovery architecture.
In addition to the above two architectures, do companies have no better choice? no! In recent years, the Storage Network based remote disaster recovery solution with Storage Gateway as the core not only combines the advantages of the host-type and storage system-based remote disaster recovery architecture, but also breaks the remote disaster recovery in one fell swoop. The construction of high-cost ice, which has become the most popular remote disaster recovery technology at this stage.
The storage network type remote disaster recovery architecture is a storage area network (SAN) between the front-end application server and the back-end storage system. A layer of storage gateway is added. This gateway is different from the network gateway we know. It combines IPStor storage. Management software and a dedicated manager, the front end connects to the server host, and the back end connects to the storage device. Its role is like a traffic police in the storage network, and all I / O is controlled and managed by it.
Storage network-based remote disaster recovery: compatible
Taking FalconStor's IPStor as an example, the operation mode of IPStor remote disaster recovery is to combine the host-side agent (Agent) and snapshot function (TImeMark) to create a snapshot copy of the host data and copy it to the remote disaster recovery through the network. Site. Because IPStor adopts the mode of copying only the changed data, the required bandwidth is low, and the connection between the local end and the disaster recovery end can use the relatively inexpensive IP network.
The advantages of storage network-based remote disaster recovery can be said to be the combination of the two aforementioned architectures. Since data replication is performed through the storage gateway, the application server only needs the database to execute the agent program, and its performance impact is very low compared to the host-type remote disaster recovery. In addition, through the virtualization technology of the storage gateway, the front-end heterogeneous platform server and the back-end storage devices of different brands can be integrated. The local and disaster recovery devices do not need to be configured in pairs, and users can establish them at the remote end according to RTO and RPO. Complete hot backup center, take over business operation immediately when a disaster occurs on the local side; or adopt a warm station configuration that only installs storage devices on the disaster recovery side, first protect the integrity and security of data, and then repair after the local side is completed Perform recovery.
In addition to the above-mentioned features that do not occupy the running resources of the host and do not have the limitations of the storage platform, cost is the biggest advantage of storage network-based remote disaster recovery.
First of all, there is no need to replace the original IT infrastructure when constructing, just add a storage gateway to the original storage area network. The local host and storage device can be any brand, and the host and storage device on the disaster recovery side do not need to be local. At the same end, users can even use lower-level storage systems (such as SATA disk arrays) at the disaster recovery end. According to statistics, investment costs can be saved by as much as 30% to 50%! For those customers who have an eager need to build remote disaster recovery and have a very limited IT budget, storage network-based remote disaster recovery is undoubtedly the best choice.
Secondly, for database-specific agents, storage network-based remote disaster recovery ensures that the database has complete disaster recovery and startup capabilities, without worrying about the phenomenon of failure to start.
More importantly, multi-point snapshots and other value-added functions on the stored data can enable the instantaneous recovery of all kinds of data. It is really the most complete disaster recovery system, which covers disasters The range of resistance is far beyond the aforementioned various methods. In this disaster recovery system, the construction of disaster recovery is no longer difficult, and disaster is no longer extremely terrible.
Remote disaster recovery based on backup tapes is currently widely used by enterprises. Managers only need to regularly back up data to offline tapes, and then transport them to remote storage through human labor. Once a disaster occurs, the IT Repair the facility, and then restore the system and data from the tape. Although this method of remote disaster recovery has the lowest cost, the operational losses caused by the long recovery time (usually requiring days or even weeks), as well as the data loss caused by the non-real-time nature of backup / recovery time, have obviously exceeded The range that can be tolerated by enterprises today. The ideal remote disaster recovery architecture should be based on an unlimited wide-area network, and use remote data replication (Remote ReplicaTIon) technology for online (on-line) remote disaster recovery, so as to meet the requirements of enterprise RTO and RPO.
Through the connection of the network, it is not difficult to copy the local data to remote storage, but under a complex IT architecture, it is not easy to achieve success. After several years of technological development, online remote disaster recovery has produced the following two types: host-based (Host based) and storage system (Storage based).
Host-type remote disaster recovery: very limited
Host-type remote disaster recovery, in simple terms, uses TCP / IP through data replication software (such as Veritas Volume Replicator) installed on the server, or data replication / disaster recovery tools (such as Oracle Flash Recovery Area) provided by applications The network connects to the remote backup server and performs data replication in an asynchronous manner.
The advantage of host-type remote disaster recovery is that in an environment with fewer servers, the cost is lower, users do not need to replace too many existing system architectures, and do not have to worry about the compatibility of the back-end storage system, only need to pay Software licensing fees and hardware equipment costs at the disaster recovery end are sufficient. However, in an environment with a large number of servers, the complexity of management will increase, and the overall input cost will also increase. Another disadvantage of it is that the software is installed on the application host, and it will consume the running resources of the host when running. If the hardware level is not high, it may affect the application.
Storage system type remote disaster recovery: expensive
Storage system type remote disaster recovery, as the name suggests, is based on the storage system (fiber disk array, NAS) mode. Through the built-in firmware (firmware) or operating system of the storage system, through the IP network or DWDM, Fibre Channel and other transmission interface links, the data is copied to the remote side in a synchronous or asynchronous manner. Well-known remote disaster recovery solutions for storage systems include SRDF, TrueCopy, PPRC, etc.
Compared with the host-type remote disaster recovery, the advantage of the storage system-based remote disaster recovery is that it separates the data from the operation, and has little impact on the operating resources of the host system. In addition, because most of the operating mechanisms use mirrors to copy data and use cache memory to speed up I / O access, the data difference between the two ends is relatively small, and the storage system itself has certain fault tolerance and certain Operating performance and reliability.
The biggest limitation of storage system type remote disaster recovery is its expensive construction cost. Since users must configure two sets of the same storage system at the local end and the disaster recovery end, not only is the procurement cost high, but also subject to a single equipment manufacturer, the future scalability is bound to lack flexibility. In addition, if the Fibre Channel storage system is to build remote disaster recovery, one FC-to-IP adapter must be installed on the local end and the disaster recovery end, and the hardware cost will exceed US $ 50,000, plus the network bandwidth cost. The overall cost input will be staggering. If the enterprise does not undergo a careful evaluation before installation, building a storage system type remote disaster recovery is likely to cause a black hole in IT expenditure and increase the financial burden. In addition, the storage system type disaster recovery method has a great defect for the consistent disaster recovery of the database, and it is not applicable to the multi-point to one-point disaster recovery architecture.
In addition to the above two architectures, do companies have no better choice? no! In recent years, the Storage Network based remote disaster recovery solution with Storage Gateway as the core not only combines the advantages of the host-type and storage system-based remote disaster recovery architecture, but also breaks the remote disaster recovery in one fell swoop. The construction of high-cost ice, which has become the most popular remote disaster recovery technology at this stage.
The storage network type remote disaster recovery architecture is a storage area network (SAN) between the front-end application server and the back-end storage system. A layer of storage gateway is added. This gateway is different from the network gateway we know. It combines IPStor storage. Management software and a dedicated manager, the front end connects to the server host, and the back end connects to the storage device. Its role is like a traffic police in the storage network, and all I / O is controlled and managed by it.
Storage network-based remote disaster recovery: compatible
Taking FalconStor's IPStor as an example, the operation mode of IPStor remote disaster recovery is to combine the host-side agent (Agent) and snapshot function (TImeMark) to create a snapshot copy of the host data and copy it to the remote disaster recovery through the network. Site. Because IPStor adopts the mode of copying only the changed data, the required bandwidth is low, and the connection between the local end and the disaster recovery end can use the relatively inexpensive IP network.
The advantages of storage network-based remote disaster recovery can be said to be the combination of the two aforementioned architectures. Since data replication is performed through the storage gateway, the application server only needs the database to execute the agent program, and its performance impact is very low compared to the host-type remote disaster recovery. In addition, through the virtualization technology of the storage gateway, the front-end heterogeneous platform server and the back-end storage devices of different brands can be integrated. The local and disaster recovery devices do not need to be configured in pairs, and users can establish them at the remote end according to RTO and RPO. Complete hot backup center, take over business operation immediately when a disaster occurs on the local side; or adopt a warm station configuration that only installs storage devices on the disaster recovery side, first protect the integrity and security of data, and then repair after the local side is completed Perform recovery.
In addition to the above-mentioned features that do not occupy the running resources of the host and do not have the limitations of the storage platform, cost is the biggest advantage of storage network-based remote disaster recovery.
First of all, there is no need to replace the original IT infrastructure when constructing, just add a storage gateway to the original storage area network. The local host and storage device can be any brand, and the host and storage device on the disaster recovery side do not need to be local. At the same end, users can even use lower-level storage systems (such as SATA disk arrays) at the disaster recovery end. According to statistics, investment costs can be saved by as much as 30% to 50%! For those customers who have an eager need to build remote disaster recovery and have a very limited IT budget, storage network-based remote disaster recovery is undoubtedly the best choice.
Secondly, for database-specific agents, storage network-based remote disaster recovery ensures that the database has complete disaster recovery and startup capabilities, without worrying about the phenomenon of failure to start.
More importantly, multi-point snapshots and other value-added functions on the stored data can enable the instantaneous recovery of all kinds of data. It is really the most complete disaster recovery system, which covers disasters The range of resistance is far beyond the aforementioned various methods. In this disaster recovery system, the construction of disaster recovery is no longer difficult, and disaster is no longer extremely terrible.
Hifi Earbuds,Best Hifi Earbuds,Hifi In Ear Headphones,Best Hifi Earphones
Dongguang Vowsound Electronics Co., Ltd. , https://www.vowsound.com