Emerging Technologies and their Chances for Success

There was a lot of excitement at the Interop conference in Las Vegas last month, in part because of the dramatic change going on in the industry.

Many of the educational sessions, for example, focused on comparing and contrasting emerging technologies. Speakers in one session discussed replacing the Spanning Tree Protocol (STP) in datacenter LANs with options that included Transparent Interconnection of Lots of Links (TRILL), Shortest Path Bridging (SPB) and Multi-Chassis Link Aggregation Group (MC-LAG).  

Speakers in another session looked at ways to move virtual machines (VMs) between physical servers using two different emerging protocols:  Virtual eXtensible LAN (VXLAN) and Network Virtualization using Generic Routing Encapsulation (NVGRE). In each case, when a speaker advocated for a given technology, their approach was always to try to demonstrate that the technology was better than its alternatives. That raises an important question: So what?  

The best technology does not always win in the marketplace as the old VHS-vs.-Betamax videocassette recorder war proved.  In our industry there are numerous examples of very powerful technologies that nonetheless died early. If you've been around awhile, you might recognize a few: ATM25,  Switched Multimegabit Data Service (SMDS) and application service providers (ASPs).

What Problem Gets Solved?

One of the first questions that has to be asked of any new technology is whether it solves a problem for which IT organizations will pay.  It is important to realize that the question is not just whether a technology can enable IT organizations to do something interesting; its relevance comes down to resource allocation.  There needs to be a justification for IT organizations to spend the required resources to evaluate, trial and implement the new technology.  That justification has to be compelling enough to cause a typically resource-constrained IT organization to invest in that technology and not another.

The second question to be asked is related to the first one: "If this is not the first technology to solve that problem, is it notably better than what already exists?"  To illustrate how this plays out in the marketplace, consider two very different scenarios: the adoption of LAN switching in the late 1990s and early 2000s, and the adoption of VoIP in the early to mid 2000s. When LAN switching was first introduced, the shared-medium LANs of that era were reaching exhaustion and there were few plans to significantly increase the capacity of them.  For example, Fiber Distributed Data Interface (FDDI), a common backbone LAN in the mid to late 1990s, ran at 100Mbps, and there were no plans to increase that speed.

IT organizations had little choice but to move away from shared, contention-based LANs and to adopt switched LANs, which dedicated full bandwidth to each port.  As a result, switched LANs went mainstream in just a few years.  That was not the case with VoIP.  

When VoIP was first introduced, most organizations had a fully functioning voice network that few IT managers stayed up late at night worrying about. When VoIP was first deployed, it was not a notably better solution to transporting voice calls. The good news is that VoIP functionality and the problem that VoIP was intended to resolve - enabling the convergence of voice and data networks into just one network to operate and manage - both evolved over time.  As a result, VoIP has been very successful in the market, but that success took close to a decade.

The Backer Factor

Another factor that drives the success of a new technology is the base of vendors that are strongly advocating for each of the competing technologies.  VXLAN and NVGRE are a great example of where there currently are camps of vendors arguing for alternative technologies.  

VXLAN is backed by some very influential players including: VMware, Cisco, Citrix and Red Hat.  NVGRE is also backed by some very influential players, including: Microsoft, Intel, HP and Dell.  While a couple of players are proponents of both, such as Arista and Broadcom, for the most part, each camp is pushing for their solution and avoiding developing the other.

It is always good for IT organizations to develop a deep understanding of emerging technologies.  However, most IT organizations want to implement technologies that will be successful in the marketplace, so they need to take into consideration the fact that the winner in the marketplace is not always the "best" technology. Sometimes the technology with the strongest set of backers is the one that succeeds.