It's highly likely that, if you follow the comings and goings of storage networking technology, then by now you've at least heard of Fibre Channel over Ethernet (FCoE). For the uninitiated, FCoE is a method of encapsulating the packets that would normally flow across a Fibre Channel storage network (SAN) for transmission over Ethernet link. Think 10 Gbps Ethernet for practical purposes. However, unlike other FC-to-Ethernet encapsulation methods like FCIP and iFCP, FCoE lives within the same OSI layer as IP, enabling enhanced performance, lossless frame transmission, and some other goodies.
For large enterprises operating distinct Fibre Channel storage-area networks (SAN) and Ethernet data networks, FCoE's potential benefits and cost savings are promising, even though the technology requires new hardware.
“Fibre Channel over Ethernet allows IT organizations to sim¬plify their datacenter network topology by reducing the num¬ber of network interfaces on their servers and simplifying the network access layer. It protects existing SAN investments by integrating seamlessly with existing Fibre Channel networks and storage devices, and it allows organizations to integrate FCoE-enabled storage systems as they become available,” says Martyn Molnar, Regional Sales Director, NetApp.
Rabih Dabboussi, Director of Systems Engineering, Cisco, offers another perspective on the primary benefits of FCoE: “The evolving business and technology requirements necessitate the need for high-density, high-speed, and low-latency consolidated communications network that enable data center scalability, eliminates bottlenecks and improves manageability. FCoE is aimed at providing these benefits by eliminating the need for separate networks such as Ethernet and fiber-channel. FCoE also eliminates the bottleneck effect caused by network-interface-cards (NICs) in most servers today and allows IT departments to unleash the power of the ever-growing computing and memory capacity of their servers.”
Adopting FCoE lets a company consolidate server hardware and cabling at the rack level. That’s the first step in moving to FCoE. The more consolidation can take place in the core switching fabric, and that reduces cost even further. Ultimately, storage arrays will support FCoE natively, making end-to-end FCoE possible.
Industry watchers say FCoE will happen in three phases: FCoE server enablement, which means the ability to have Converged Network Adaptor (CAN). “Today, data centres supporting mission-critical operations use redundant Ethernet Network Interface Cards (NICs) in their servers and redundant pairs of switches at each layer in their network architecture. A new class of CNAs help to further simplify FCoE deployment, allowing FCoE to be incor¬porated without disrupting current datacenter management practices, software, or the roles of network and storage administrators.
CNAs present both an Ethernet interface and a Fibre Channel initiator to the server, allowing the operating system to see two physical devices. CNAs can be implemented in software, or in hardware. This technology makes the existence of the converged network transparent to the operating system and applications, allowing both storage and network adminis¬trators to manage their respective domains just as they do today. Consistent management helps ease FCoE deployment while reducing operating expenses. Phase two involves FCoE server proliferation with phase 3 translating into FCoE arrays and SANs.
Tall orders for FCoE
The promise of FCoE lies in a merged data and storage network fabric, but making the integration happen isn't easy and shouldn't be done without careful planning. Typically in an enterprise, network and storage management teams don't work together. That has to change with FCoE. Another challenge is that FCoE doesn’t run on standard Ethernet and required Converged Enhanced Ethernet, an advanced version of Ethernet aimed at data centres.
“The great thing about FCoE is that it allows a slow migration from FC to Ethernet. It’s possible to expand or replace part of your FC network with Ethernet switches, allowing you to make an orderly transition from one networking technology (FC) to the other (Ethernet) as it makes sense. As storage traffic requires the new lossless Ethernet, the 10GbE transport still requires Link Layer multipathing and multihop capabilities. Such features are currently under development, and should become available later in 2010. These capabilities will enable the deployment of larger FCoE networks, which will expand the reach of FCoE beyond access layer server connectivity and I/O consolidation,” says Molnar.
Should you bite? Is it too early to make investments in FCoE now? “Not at all, as now FCoE draft has been finalised and the standard ratified, plus we see most major data centre vendors adopting FCoE as the key enabler to next-gen data centre architecture. We believe that FCoE will be the basis for the unified fabric in all data centre environments,” says Dabboussi.
NetApp’s Molnar says the decision is up to individual companies. “Each IT organization should look closely at their situation. For those with large investments in Fibre Channel who have no need for upgrades in the next couple of years, the best bet is probably to do nothing. However, if upgrades are being planned in that time frame, then they should take a serious look at FCoE. Given that the current FC switch vendors will be migrating most of their customer base to Ethernet, at some point companies will probably end up making the switch.”
Will FCoE cost more than FC? Yes, at least in the beginning. With both FC and Ethernet vendors lining up behind the FCoE standard, what will happen to FC? Dabboussi says FC and FCoE will co-exist. Will the transition to 16Gb technology still occur, or will FCoE be in the driver's seat? “As it stands right now, 16Gb FC is planned for 2011. In a recent press release the FCIA stated that it “strongly” supports the development of 16Gb FC and it also supports FCoE. We at NetApp, believe 16Gb FC will become reality, but a larger question will be its adoption rate relative to FCoE. That remains to be seen,” says Molnar.
