PREVIOUSLY: performance and reliability
In the first part of our series on Fibre Channel and iSCSI storage area networks (SANs) for SMBs, we discussed the reliability and performance of each. In this part, we will discuss the cost and scalability considerations to help you decide which one is best for you.
When all out performance isn't required, ISCSI clearly has a leg up when it comes to holding down cost. iSCSI leverages the knowledgebase of existing IP network administrators, possibly reducing the staff required to support the storage infrastructure. iSCSI also offers the opportunity to reduce the cable count in low I/O situations by routing storage and general IP over the same NIC port.
Keep in mind though, if iSCSI is expected to compete with 4 Gb Fibre Channel performance, 10 GbE network ports dedicated network segments and iSCSI HBAs all cost about the same as their Fibre Channel counterparts. Finally, data storage manufactures understand the cost-conscious SMB marketplace and usually price iSCSI arrays more attractively than similar or larger Fibre Channel arrays.
iSCSI versus Fibre Channel: Which is more scalable?
iSCSI is infinitely more flexible than Fibre Channel. Fibre Channel initiator to target distances are limited by the allowed length of a fiber segment and flat networks. Because iSCSI relies on TCP/IP, which happens to power the global internet, distances are only limited by an application's tolerance for latency. iSCSI is also more flexible because it rides on the very open and resilient TCP/IP standard. As a result, iSCSI initiators, networks and targets have far fewer interoperability challenges than Fibre Channel.
Both iSCSI and Fibre Channel are viable technologies and there is considerable overlap in their sweet spots. Fibre Channel has proven to be reliable, scalable and fast while iSCSI deploys quickly with extreme flexibility at a relatively low cost point.
This chart will help you determine which protocol is best for your organisation. Select three of the five characteristics that are most important and count the number of wins for each protocol. The one with the highest score is probably the best choice.
About the author: Brian Peterson is an independent IT infrastructure analyst. He has a deep background in enterprise storage and open-systems computing platforms. He has consulted with hundreds of enterprise customers who struggled with the challenges of disaster recovery, scalability, technology refreshes and controlling costs.