| Home > How are switches changing the way SANs are designed and managed? | |
| FAQ:: |
|
||
First, decide on your needs objectives. The question has become blurred because switches are getting larger on the whole. The difference between a switch and director used to be the number of ports, the performance, the supported protocols, the modular architecture of a director and the scalability. Today, we have switches that are 64 ports that offer a level of modularity and redundancy along with multiple protocol support. So, the traditional distinctions are fading. I'd suggest taking a closer look at needs that consider ports (e.g., Fibre Channel (FC) or FICON, speeds (e.g., 1 Gbit, 2 Gbit or 4 Gbit), the need to oversubscribe ports, segmentation to isolate traffic workloads, distance requirements, scalability into the future and resiliency/redundancy requirements and so on. Then look at vendor offerings and consider products [switch or director]. Don't start with "director vs. switch," but instead consider the product that best fits your needs -- keeping future growth and needs in mind. A word about intelligent switches, which typically support storage virtualization tasks: Separate the discussion from ordinary switches and directors. Intelligent switches are not the same as directors and vice versa. Today, you need to add a blade to a director to support "intelligence," such as virtualization, you're more likely to see that functionality deployed as a specialized blade in a dedicated appliance, though this will probably change with the next generation of blades. However, there are dedicated "intelligent" switches with emphasis on virtualization and other high-end features rather than switching. Many of those have fallen to the wayside except for purpose-built SAN appliances still in the market. Listen to the SAN FAQ audiocast here. Go back to the beginning of the Storage Area Network FAQ Guide.
'); // -->
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||
|
||||||||||