The idea of Software-Defined Networking (SDN) has been around since the late 1980s/early 1990s, when it was used to describe the telecom industry’s shift from strictly private voice networks to logically constructed private voice networks that were physically hosted on the public telephone infrastructure. That shift laid the groundwork for Virtual Private Networks (VPN).
Today, SDN is re-emerging in broader but analogous discussions around cloud computing and labor automation. Here, SDN refers to a physical compute resource that is logically constructed (via virtualization) to be hosted in the cloud. Likewise, when labor is automated the “labor code” can ride on that same cloud. The result: you can connect to the cloud via the network to leverage both the virtual computing and the virtual labor resources.
In this context, SDN merges the cloud with the network and says they are one in the same. More accurately stated, the network incorporates the cloud into the network operating system. When you plug into the network you get all of the functionality available on the cloud – rather than just a fat pipe. This is the utility computing vision that Nicholas Carr laid out in the book The Big Switch.
At this point, it’s mostly just that – a vision, one that exists primarily on PowerPoint presentations and is available only in limited custom rollouts. The first true waves will focus on demonstrating the value of very high speed access/egress into an intelligent network (cloud) that logically delivers what today’s infrastructure and labor deliver in a physical sense. Needless to say, these are radical changes.
While SDN is still primarily the stuff of futuristic visionaries, it’s not too early for businesses to start exploring the journey of convergence between network evolution, cloud computing, and labor automation. All of these trends are moving and developing rapidly, albeit on separate tracks. Ultimately, however, after the inevitable security, regulatory, political and other speed bumps and detours along the way are cleared, these three trends will converge. When that happens, the IT services landscape will be fundamentally transformed – from the way that commercial contracts are negotiated and written to the manner in which services are delivered.
In addition to practical considerations, there are some philosophical questions to consider; namely, what does the replacement of physical machines and flesh-and-blood people portend for the future?
The utopian point of view holds that labor automation will drive down the costs of delivering services and create heretofore unseen value. Displaced workers will be retrained to acquire new skills and migrate to new areas; the economic pie will grow to the benefit of all.
From the dystopian perspective, wealth will increasingly concentrate into the hands of a few providers and business owners. More workers will chase fewer jobs in a losing competitive battle against low/no cost software platforms. Income levels and living standards will decline as income disparity expands.
The reality will likely be a mix of utopian and dystopian outcomes. The chart at left shows US workforce participation rates dating back to 1948; the steady decline in the curve begins around the year 2000. We’re familiar with the loss of manufacturing jobs and the struggles of workers seeking to acquire new skills and find new employment. At the same time, manufacturing automation has benefitted all consumers by decreasing the cost and improving the quality of many products.
Will the same thing happen in the IT industry? Will this trend expand to other knowledge work? It already is and we are seeing both positive and negative effects.
The challenge the industry faces is to find ways to embrace this inevitable progress and deploy it to maximize the utopian impact and minimize the dystopian. The storm of labor automation is coming. Prudent companies and workers are thinking about how to prepare to capitalize on the trend, rather than waiting to become victims. Do that and you can end up on the utopian side of the market.0