Join Telegram Study Group ▷
SPOTO has many excellent professional teams and will provide you for several specific technologies. The following is a good example to show SPOTO’s training capacity.
The two technologies of remote transmission are different: the performance of the fiber network topology is undoubtedly the first choice, but the price based on IP or the use of the existing network topology is more economical, want both fish and bear's paw A comprehensive consideration is needed to take advantage of high-speed IP networks and existing communication links.
The communication link between the local and remote ends of the disaster recovery is an important factor to consider in the remote disaster recovery solution. The biggest impact is in terms of TCO, which is due to the fact that communication links are usually the most expensive overhead in remote disaster recovery scenarios.
Remote transmission requirements
Distances Prevention of man-made disasters such as floods, earthquakes, and other natural disasters and terrorist attacks requires a distance of tens of kilometers to hundreds of kilometers between the local data center and the disaster recovery data center. The general synchronization center requires about 100 kilometers, and the asynchronous center can reach several hundred kilometers or even thousands of kilometers.
Performance The two most basic measures of any communication technology capability are:
Bandwidth - The total amount of data that can be moved over a link over a given period of time. According to the SLA (Service Level Agreement) requirements, if the disaster recovery center only needs to meet the remote tape backup function, the bandwidth requirement is < 50Mbps; if the data center integration function is to be implemented, the bandwidth requirement is 50Mbps to 1Gbps; and the remote disk image is required. The required bandwidth is above 1 Gbps.
Latency—The total amount of time it takes to move data from one end of the link to the other. The delay caused by data transmission between the local and remote ends of the disaster recovery mainly includes two parts: one is the delay caused by the distance between the two places, and the other is the delay caused by the link bandwidth between the two places.
Remote disaster recovery topology
From the perspective of topology, there are three main types of network storage technologies: DAS, NAS, and SAN. The remote disaster recovery solution generally uses the latter two.
NAS (Figure 1) is a special dedicated data storage server with embedded file service management tools that provide cross-platform file sharing. It is based on LAN (Local Area Network), uses TCP/IP protocol for communication, and uses file-level I/O for data transmission.
Figure 1
NAS (Network Attached Storage) topology
The SAN (Figure 2) is actually a dedicated network that is connected to the LAN through a gateway device, independent of the traditional LAN. SAN adopts network-oriented storage structure, data processing, and data storage separation, flexible addressing, long-distance data transmission, high I/O performance, high storage device utilization, and high user sharing. A new storage architecture.
Figure 2
SAN (Storage Area Network) topology
Remote transmission technology design
Protocols How a remote disaster recovery solution uses communication links is just as important as the link itself. Protocols have the ability to adjust bandwidth utilization, which is a significant advantage of any remote disaster recovery solution. This feature allows the user to specify how much bandwidth each application can use, that is, to allow a single link to be shared with multiple applications, each of which selects bandwidth based on its own needs. Moreover, the protocol can also automatically adjust to the current state.
The data transmission protocols used in the existing remote disaster recovery solutions mainly include FCP (Fiber Channel Protocol), FCIP (Fiber Channel over IP), and iFCP (Internet Fiber Channel Protocol). Protocols such as channel protocol) and iSCSI (Internet SCSI).
FCP (Fibre Channel Protocol)
The Fibre Channel Standard (FCS) defines a high-speed data transmission mechanism for connecting hosts and storage devices. The Fibre Channel protocol is the SCSI interface protocol on Fibre Channel. The Fibre Channel network topology supports three architectures: point-to-point, arbitrated loop, and switched fabric. With a transfer rate of 10 Gbps, optical and dielectric support, and a wide range of interface command sets including IP, SCSI, IPI, HIPPI-FP, and audio/video. If fiber is used as the transmission medium, the distance between the local and the remote can reach 10 kilometers.
FCIP (IP-based Fibre Channel Protocol)
The FCIP-based FCIP protocol encapsulates Fiber Channel commands and data using TCP/IP packets and transmits Fiber Channel commands and data over an IP network. The FCIP protocol can connect individual isolated Fibre Channel SANs through an IP network to form a unified storage area network. Therefore, it can be used to overcome the distance limitation factors existing in Fibre Channel and connect SAN islands beyond the distance supported by Fibre Channel.
iFCP (Internet Fibre Channel Protocol)
The Internet Fibre Channel Protocol (iFCP) is a gateway-to-gateway protocol that provides Fibre Channel communication services for fiber-optic devices on a TCP/IP network. iFCP is able to use the congestion control, error monitoring and recovery functions provided by TCP. The primary goal of iFCP is to enable existing Fibre Channel devices to interconnect and network at wire speed over IP networks, where IP components and technologies replace Fibre Channel switching and routing fabrics.
iSCSI (Internet SCSI Protocol)
iSCSI (Internet Small Computer System Interface) is a standard for data block transmission based on the IP protocol. It was initiated by Cisco and IBM and is strongly supported by IP storage technology advocates. The main function of iSCSI is to encapsulate and reliably transfer large amounts of data between the host system (initiator) and the storage device (target) on the TCP/IP network. Compared with the previous network access storage, the emergence of iSCSI solves the problems of openness, capacity, transmission speed, compatibility, security, etc. Its superior performance and low cost make it pay attention to the market from the date of release. With favor.
Link failure handling
How a remote disaster recovery solution handles the temporary failure of a communication link is another factor in evaluating whether the solution meets commercial needs.
There are two general scenarios for implementing link failure handling: I/O logging and pointer logging. For the I/O logging scheme, if the link fails, each I/O will copy a log file as it arrives from the host. When the link recovery is available, the log files are transferred to the remote site. For a pointer recording scheme, the source LUN is broken down into blocks; if the link fails, a pointer table is formed which shows which blocks have been modified. When the link is restored, the modified block is retransmitted to the remote location.
How do fish and bear's paws have both?
Remote disaster recovery systems generally require distances of more than 100 kilometers, and this distance requirement makes transmission the most critical factor in disaster recovery systems.
SAN is currently recognized as the best data transmission method. It is divided into FC SAN and IP SAN according to the basic network. The FC SAN uses an all-fiber channel to meet the high-performance requirements of remote data transmission. However, the cost of deploying an FC SAN-based remote disaster recovery system is extremely expensive and affordable for non-SME users. And Fibre Channel technology has its fatal weakness: Fiber can guarantee an effective transmission range of only 10 kilometers, which is not enough for remote disaster recovery systems. Although the method of physically increasing the SFP signal strength and using the SAN over SONET (SDH) method to enhance the data transmission performance of the remote disaster recovery (over 100 km) has been proposed, the deployment cost of the disaster recovery solution is further improved. The transmission of data through the IP protocol in the remote disaster recovery system is the cheapest data transmission method. This technology can achieve remote disaster recovery to some extent, but due to limitations of bandwidth and reliability of the IP network, The performance of data transfer is low and the data recovery time required is longer.
Protocols such as iSCSI, FCIP, and iFCP can overcome the shortcomings of Fibre Channel technology and IP technology, and combine their advantages to provide a feasible solution for obtaining the optimal cost performance of remote disaster recovery system. There is no performance problem with an IP SAN implemented by technologies such as iSCSI and the integration of existing FC SANs and NAS to build a remote disaster recovery system. It can utilize existing disaster recovery system facilities or low-cost iSCSI SAN deployment. The new disaster recovery system can be cost-effective for a variety of users.
With the promotion and popularization of 10 Gigabit Ethernet, I believe that the era of long-range disaster-tolerant fish and bear's paw is not far from us.
