I was sharing a few tweets with @Stu from Wikibon and @Jonisick about the newly ratified 16Gb Fibre Channel standard. The question was raised, “Great, when will 16Gb be taking over?” Obviously, this is a multi-part answer. The pace of Fibre Channel transitions has definitely slowed down with the shift from 4Gb to 8Gb. From 1Gb to 2Gb and 2Gb to 4Gb, the change took about 24 months, primarily because the cost was the same from one transition to the next. However, with the transition from 4Gb to 8Gb, we saw a non-trivial cost bump for the 8Gb optics. This means that we are over two years out from the first 8Gb shipments (February 2008) and most Host Bus Adapter (HBA) companies have not reach 50% of 8Gb port shipments. According to Seamus Crehan of Crehan Research, we should see the crossover point between 4Gb and 8Gb sometime in the first half of CY2011.

So what does this mean for 16Gb Fibre Channel? First, let’s look at where 16Gb is in the development cycle. We now have an approved standard and some early adopter companies, such as IBM, with whom we announced a 16Gb design win last fall before the standard was finalized.

• Step One: Standard Approval. As we all know, development work happens pre-standard and it is well underway with all of the Fibre Channel vendors (optics, HBAs, switches, target devices) working toward 16Gb.
• Step Two: Industry Plug Fests. We expect that we will see the first industry interoperability plug fests with the University of New Hampshire, in the first half of CY2011, about the same time 8Gb reaches 50% of ports shipped.
• Step Three: Public Demo, You can bet that during some tradeshow in Q2-Q3, we will see the first end-to-end 16Gb demo. Probably with NetApp, since they have a standard PCI-e card slots in their array.
• Step Four: Vendor Hype. As with every new speed bump, the vendors will debate the virtues of the new vs. the old vs. the alternatives to Fibre Channel (10/40GbE FCoE, iSCSI, NAS, IB) all over again. Emulex included.
• Step Five: OEM Qualifications. I expect to see server OEMs qualify the first 16Gb cards in the second half of CY2011 and they will ship around the end of CY2011 or early 2012. Most storage vendors have moved to some form of modular host interface architecture that will accelerate time to market on the target side, but they will wait to follow the server vendors, since they need to support the cards approved for servers.
• Step Six: OSs, Hypervisors and Apps. Another thing to remember is that speed bumps are more than protocol bumps; they are the excuse/reason to add a bunch of new features and capabilities for next-generation server and storage platforms. For 16Gb, this includes things like PCI Express 3.0, new virtualization tools, SR-IOV, and other cool features that require not only hardware changes, but OS, hypervisor and application changes to evolve the entire ecosystem.
• Step Seven: IT Managers Evaluation. Of course, this is the phase of the market that really matters. Will people buy it or ignore it? This will mostly happen at the end of CY2011 and early CY2012 and then we will see a transition that probably resembles the 4Gb to 8Gb transition over a multi-year time horizon.

I am sure some folks might take this as anti-Fibre Channel, but it is not; I have said the same things about network convergence (FCoE) and 10GbE adoption. This should not be news to industry vets like @Stu and @Jonisick, but it felt like a subject worthy of a blog. Hopefully it will stir up added conversations about market transitions and help continue the vendor debates and prognostications about what protocols will win in the market.

Posted in Uncategorized | No Comments