There are a number of different methods and protocols for linking geographically distributed storage area networks.
BY TOM PETROCELLI
One of the most promising areas of storage technology today is in SAN-to-SAN connectivity. Extending storage area networks (SANs) beyond the local fabric, also known as SAN transport, will greatly improve data availability, disaster avoidance, and storage consolidation. This practice allows established enterprises to integrate geographically dispersed storage resources and is a critical enabling technology for e-business. Long-distance SAN-to-SAN connectivity is also important for storage service providers (SSPs) since it allows economics and business needs, rather than location, to dictate the placement of consolidated storage facilities.
Unfortunately, confusion exists in the marketplace regarding SAN-to-SAN connectivity. Several competing technology camps are claiming their approach is the "best." Multiple protocols-some based on open standards and some proprietary-and a range of opinions on what is considered the "best" option for SAN connectivity are sending mixed signals to the IT community. This uncertainty threatens to delay deployment of this useful technology to the detriment of business operations.
An open and inclusive model is required to solve this problem. A multi-service access approach for SAN-to-SAN connectivity addresses the multiplicity of applications, needs, and technology tools that can be leveraged to make SAN-to-SAN connectivity work.
Several distinct methodologies are available for performing SAN-to-SAN connectivity. The most widely deployed method of extending SANs past local fabrics is a server-based method. Integrated solutions based on existing LAN and WAN network infrastructures, Fibre Channel over ATM and SONET, and Fibre Channel-to-T1/T3 are quickly emerging. The viability of each methodology depends on the needs of different storage applications.
Server-based approaches employ a server with host bus adapters, network interface cards, and specialized software for routing either Fibre Channel packets or SCSI blocks to a LAN for transport over an existing WAN. Server-based solutions can be deployed immediately but often complicate long-distance SAN transport. Server-based SAN transport can be costly and difficult to manage because it typically requires a high-end server with vast amounts of memory, powerful processor(s), and specialized software. Since the data transport software is not based on an open standard, this is essentially a proprietary approach.
Another approach is to deploy devices that leverage current Ethernet-based network infrastructures to provide a bridge between Fibre Channel SANs and existing LAN-to-WAN routers and switches. This method is not to be confused with "storage over IP" systems, in which all storage traffic, including SCSI block I/O, is run over an Ethernet or Gigabit Ethernet network.
The multi-service access method supports a number of different networking technologies such as T3, OC-3, Gigabit Ethernet, ATM, and OC-48.
The obvious advantage of transporting Fibre Channel packets over existing LANs and WANs is that almost all of the infrastructure is already in place. In addition, the IT community is very familiar with its current infrastructure, resulting in faster implementation and easier management. The major disadvantage is congestion. Running block I/O traffic with other traffic can overload the primary network and, during periods of heavy usage, prevent storage traffic from arriving in order and on time. In addition, data must be routed through multiple switches and routers before it enters the WAN, compounding latency problems and complexity.
For these reasons, Fibre Channel-to-Gigabit Ethernet is likely to be restricted to occasional access, emergency fail-over from a dedicated storage WAN link, or to periods when network traffic is light (e.g., after business hours) and where moderate to low bandwidth is required.
OC-48 and SONET
In many cases, a dedicated connection to remote storage resources is preferred, especially where large amounts or constantly available block I/O traffic is expected to flow between geographically dispersed SANs. OC-48 and SONET are effective technologies to provide SAN-to-SAN connectivity because both combine high throughput (up to 2.5Gbps) with multi-point service.
Disadvantages of SAN transport via OC-48 and SONET include its relatively short range (restricted to metropolitan area networks), high cost, and lack of availability to many enterprises. OC-48 and SONET are good choices when I/O-intensive applications are run over short distances. However, storage applications must be mission-critical to justify the expense of dedicated solutions.
T3 and OC-3
The most common networking technologies that support SAN-to-SAN applications are T3 and OC-3. While transmission speeds are low compared to Fibre Channel, dedicated T3 or OC-3 lines are sufficient for applications that have moderate bandwidth needs and do not require continuous remote access. In addition, if an application is supported by SAN-to-WAN routers that transfer Fibre Channel data over several T3 or OC-3 links, then low to moderate bandwidth connections can be an effective compromise between performance, distance, and cost.
The obvious drawback of T3 and OC-3 connections is the relatively slow data-transmission speed. At 45Mbps and 155Mbps, respectively, T3 and OC-3 provide much lower bandwidth than OC-48 does at 2.5Gbps. Still, the cost and distance profile of T3 and OC-3 service makes aggregation of bandwidth over these links more attractive, especially for applications that do not require full gigabit speeds.
Open vs. proprietary protocols
Another important element of SAN-to-SAN connectivity is transport protocols, an area in which open standards-based systems are competing with proprietary protocols.
Many vendors are backing the development of open protocols being put forth by the major standards bodies. Examples include FC over IP, which is currently under discussion by the Internet Engineering Task Force, and Fibre Channel Back Bone, which is under revision by the American National Standards Institute T11.3 committee. Both proposed standards represent an attempt to develop an open protocol for transporting Fibre Channel SANs over IP, ATM, or SONET networks.
The second approach is to use protocols specific to a particular vendor's product. Of course, relying on vendor-specific solutions essentially yields a closed system-a common problem in the server-based approach and to early SAN/WAN routers. Closed systems lock IT managers into a specific vendor's solution, which may not service all of their needs effectively, and offers no protection against technology obsolescence. Some vendors are developing protocols with the intention of establishing de facto standards, adding to the confusion and IT professionals' reluctance to deploy a solution.
Multi-service access approach
Faced with so many competing approaches, vendors, integrators, and end users are challenged by questions such as: Which type of connection should I use? What protocol is the right one? Whose approach will dominate? Many IT professionals with responsibility for multiple storage applications believe they have no choice but to deploy products from different vendors, many of which are not interoperable.
An alternative to the multi-vendor solution is the multi-service access approach. Providing a variety of WAN, LAN, and MAN interlinks that support different protocols, multi-service access offers IT managers the opportunity to make decisions based on need and budget. Multi-service access frees IT professionals from being at the mercy of a particular vendor. The application, rather than the router vendor, can now drive the connectivity planning. By deploying a variety of open protocols, IT managers can effectively and efficiently interoperate with other facilities, customers, and suppliers.
The multi-service access concept is not new. In fact, it has been a fundamental element of traditional networking for years. The multi-service access mind-set can now be applied to SAN-to-SAN connectivity to provide IT managers with the right tools for meeting the constantly escalating storage demands of their enterprise systems.
Multi-service access offers five potential advantages:
- WAN-link deployment decisions can be based on the needs of mission-critical applications. Multi-service access can also maximize applications performance.
- It supports tighter cost control because bandwidth can be purchased based on need. Controlling bandwidth costs is critical because WAN links are a recurring monthly expense.
- The multi-service access approach ensures investment protection because incremental upgrades in connectivity can be ordered with minimal changes and disruptions to the overall system. As the system requirements grow, decisions can be based on economics rather than a particular vendor's product schedule.
- Multi-service access provides efficient service across the enterprise. For example, OC-48 and SONET are not universally available. Even within a single company, WAN infrastructure varies widely. Accordingly, relying on one approach for SAN-to-SAN connectivity means that some locations will not be able to participate in an application until that service is available. If a lowest common denominator solution such as T3 is implemented, some bandwidth-intensive applications cannot be deployed.
- Multi-service access based on open protocols enhances interoperability. As SAN-to-SAN connectivity becomes increasingly important, the interoperability of equipment from multiple vendors and different connectivity options is vital. Moreover, in a time when mergers and acquisitions are commonplace, interoperability is an important part of IT planning and should be considered when deploying SAN-to-SAN applications.
The SAN-to-SAN connectivity landscape is new and fragmented. Several different schools of thought have emerged regarding the "correct" way to allow geographically dispersed SANs to interoperate. Advancements in Fibre Channel, Gigabit Ethernet, WAN infrastructure, and other key storage networking technologies are making the multi-service access approach a reality.
Tom Petrocelli is vice president of business development at Entrada Networks (www.entradanetworks.com) in San Diego, CA.