By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
Options for scaling a storage networking beyond a single fabric include:
- Establish additional independent fabrics consisting of one or more switching devices with various numbers of ports per switch.
- Interconnect fabrics with inter-switch links (ISLs) creating one large physical and logical fabric. While this approach enables any to any connectivity, it does not increase the addressing range and number of devices supported per fabric. This approach also increases the amount of management traffic moving around a fabric.
- Interconnect additional and existing fabrics with ISLs attached to a storage network router that supports fabric segmentation. The storage network fabrics are physically connected, yet logically isolated similar to how TCP/IP-based Ethernet LANs are often configured. This approach has the benefit in that it can be deployed on a local and wide area basis enabling many devices to be part of a larger storage network, yet logically isolate devices and fabric resources.
- For spanning distances, technologies including TCP/IP (IP), SONET/SDH, and optical multiplexing (CWDM/DWDM) can be used. FCIP and iFCP can be used to connect fabrics together over distances using TCP/IP based networks.
- Protocol conversion enables devices and servers outside of a fabric to access storage resources in a fabric. An iSCSI server can access resources such as disk space capacity or tape backup capability located in a Fibre Channel SAN fabric.
- Switch partitioning is a relative new capability particularly for Fibre Channel- and FICON-based switching products. Partitioning differs from segmentation in that with partitioning, a single large switch is subdivided into multiple smaller logical switches (domains). Each logical domain functions with its own fabric resources (name space) and identity including domain ID. Specific ports and internal bandwidth are allocated to the different local switches. This approach enables a physical single switch to have different logical switches be part of different fabrics. Since each of the fabrics that a logical switch can be part of is autonomous, some form in routing function would be required. The routing can be performed using an external router, be it can be implemented as part of switching devices hardware and firmware feature set.
About the author: Greg Schulz is a senior analyst with the independent storage analysis firm, The Evaluator Group Inc. Greg has 25 years of IT experience as a consultant, end user, storage and storage networking vendor, and industry analyst. Greg has worked with Unix, Windows, IBM Mainframe, OpenVMS and other hardware/software environments. In addition to being an analyst, Greg is also the author and illustrator of Resilient Storage Networks, Greg has contributed material to Storage Magazine. Greg holds both a computer science and software engineering degree from the University of St. Thomas.