FCoE has been on the IT radar screen for a few years, but there's still a lot of confusion -- and not a lot of guidance for end users -- out there around it. If a customer meets someone who can clearly articulate the strengths and weaknesses of FCoE, they may accelerate their plans, especially if by doing so they can save significant capital and operating expenses. The organization, supplier or integrator that leads this conversation and establishes himself as the trusted advisor has the best shot at leading the project.
So that's why you should become an FCoE expert. But how?
Understanding the landscape
The first step -- useful for FCoE, but also when learning any new technology -- is to understand the scope of the solution set and what role the technology plays in that set, as well as what the alternatives are. In short, make sure you understand the whole puzzle and not just one piece of it. So let's put FCoE in the context of the broader solution set it belongs to: networking infrastructure consolidation. A company interested in FCoE will be interested in reducing the number of server I/O cards and cables running through a rack of servers. (If you've been in a data center of any size lately, you know that cable management is becoming a key issue, as our article "Cable Survival Techniques" on cable management explains.)
There are two other types of products that enable networking infrastructure consolidation: I/O virtualization products and backbone replacements. I/O virtualization technologies use an I/O controller to house multiple I/O cards, and then a single cable connects into the server host, typically via a PCI-E extension card or an InfiniBand card. Companies like Xsigo, VirtenSys and Aprius are in the I/O virtualization space. The products' goal is similar to FCoE: to reduce cable count and NIC card count within the rack.
Consolidation at the backbone is also key. Companies that offer blade-based backbone replacements for smaller switches, such as Brocade and Cisco, are reporting brisk business. Like I/O virtualization products, backbone replacements aim to cut down on hardware -- specifically, to reduce the number of standalone SAN switches in the environment.
All three of these technologies -- FCoE, I/O virtualization and backbone implementation -- are complementary to one another but can be done as standalone projects. For example, an I/O virtualization switch could house a few FCoE cards that are shared to the servers in the rack, or it could use few 8GB Fibre Channel cards and a few 10 Gigabit Ethernet cards. In either configuration, a single pair of cables would run down the rack to the each server.
Backbone consolidation also lends itself to a standalone or complementary approach. Some backbone infrastructures now support FCoE blades so that an FCoE connection could run from the server directly to the backbone rather than to a top-of-rack switch. Similarly, an I/O virtualization switch at the top of the rack could run directly to the backbone blade. Each of the strategies could be implemented in phases or all at once.
While FCoE continues to get more attention in the market than I/O virtualization and backbone consolidation, it's important to bring those two other options into the discussion; it will show you have a broader understanding of the problem and are able to discuss multiple options. (You are, after all, an integrator.)
Research and in-depth training
Now you have a good idea of the scope of what needs to be learned. Next comes the actual learning process. How you go about this is really a matter of personal preference. You might choose to start out with a Web browser, researching the topic. The value of the Internet is that in seconds, more information, in almost every format, than you can ever read or watch is at your fingertips. For example, if you type "FCoE" at youtube.com, at least 20 whiteboard presentations come up, pro and con, on the subject. This is important because rather than relying on the top one or two resources, you need to gather information from a number of sources and cross-check that information to get a more accurate picture of the reality.
The great thing about the first part of the learning process is that it's free. And it may give you enough confidence that you're ready to step into the conversation. At the very least, you'll be able to assess the technology and see if it's worth investing in. If so, you'll need to spend more time and some money, either on lab equipment or training classes. In the case of FCoE or infrastructure consolidation, if your target market is midsize and larger data centers, especially those involved in server virtualization projects, the time to spend that money is now.
Assuming you take my advice, you're ready to get hands-on with FCoE. There are typically three ways to do that: Bring the product into your lab, go to a training class or get involved with a client that's considering an infrastructure consolidation project. Personally, I like to be left alone to figure the technology out. Give me the product, a 12-pack of diet Mountain Dew, and I'm ready to roll. Others will want to go to an FCoE training class first instead. The final option -- getting involved with a customer evaluation -- may seem odd to you, but I find it to be, by far, the most valuable. Seeing how other people learn and react to the technology gives you excellent data points as you begin to develop an integration strategy. I typically do this type of customer engagement at no charge. After all, I'm learning on their equipment. And with FCoE, it's hard to create a lab environment big enough that can really expose all the issues that may arise. In the customer environment, you are much closer to a production simulation and should be able to stress the infrastructure more.
Once these two learning cycles -- researching and in-depth training -- are complete, there comes a time to develop a customer engagement strategy and an implementation strategy for FCoE -- both of which I will cover in the next part of this series.
About the author
George Crump is president and founder of Storage Switzerland, an IT analyst firm focused on the storage and virtualization segments. With 25 years of experience designing storage solutions for data centers across the United States, he has seen the birth of such technologies as RAID, NAS and SAN. Prior to founding Storage Switzerland, George was chief technology officer at one of the nation's largest storage integrators, where he was in charge of technology testing, integration and product selection. Find Storage Switzerland's disclosure statement here.
This was first published in November 2009