The problem of familiarity

It’s always prudent to do a detailed analysis of benefits and costs before a major expenditure. We want to know the total cost of ownership, including one-time costs (materials, installation, training), recurring costs (licensing, maintenance, power, depreciation), and risk costs (overruns, supply chain issues, disasters). And we want to know what value we can derive from the expenditure – more customers, higher profit, operational savings, better predictability, etc.

But this analysis isn’t easy. We tend to be good at identifying benefits and costs for a solution that’s like what we’re doing today. But familiarity can cause us to overlook the hidden risks of those solutions, and stick with the present mode of operation. And we are bad at the same analysis when a technology is innovative or disruptive. Let’s look at examples of the hidden costs of current approaches, and the benefits associated with technology revolution – specifically, the move from closed and proprietary devices to open and disaggregated systems.

Hidden costs? Or just overlooked?

In our own lives, we know that eating too much, drinking too much, smoking, and lack of exercise are bad for us. But that downside is often theoretical – until it isn’t. A diagnosis of heart disease, diabetes or cancer is an unwelcome wakeup call. It shouldn’t be a surprise, but it usually is to the patient.

Likewise, we know that today’s closed and proprietary solutions have risks that can lead to costs. But a stable environment may lead us to ignore those risks until it’s too late. Here are some examples:

Industry consolidation – There’s always a risk to supplier selection. What happens if your selected supplier is acquired, and then the product is discontinued, or the price goes up? What happens if they can’t keep up with the innovation from their competitors? Commitment to proprietary devices makes it difficult to respond quickly if these scenarios happen.

Supply chain issues – Wise buyers always try to have a second source for materials. But others fail to see the risk and opt for a solution that’s unique. When pandemics or trade wars strike, they may be out of luck because they don’t have a viable substitute for the product they can’t get.

Closed solutions – Trends such as movement to the cloud, agile development and CI/CD are accelerating the development and deployment of software. But they rely on open systems that can be updated as needed. Selection and deployment of closed systems means that you’re reliant on that one supplier for all future innovation.

Selection and deployment of closed systems means that you’re reliant on one supplier for all future innovation.

When we’re looking at a new technology we may focus on the advertised features and benefits. For example, disaggregated systems can address the hidden costs described in the previous section. And that’s great in and of itself.

But sometimes there are other more subtle benefits that turn out to be very important. Let’s look at some non-obvious advantages of moving to open and disaggregated solutions – such as the following:

Reduction in inventory cost – Let’s consider the case of a telco supporting managed services such as VPNs, firewall, SD-WAN and voice services. Customers often have preferences for the supplier used for these critical services, so the telco may need to support multiple options for each service. And they will need to support multiple sizes of deployments (small, medium and large). That means supporting a large number of physical device types. And providing tight SLAs means storing multiple instances of each at numerous depots. That’s a lot of capital tied up in idle inventory. Moving to a virtualized approach means that large array of disparate devices can be replaced by just three standard servers (small medium and large) that can host any required application. And that means the inventory for each depot is radically reduced.

Easing international deployments – Countries like Brazil have onerous requirements for certifying equipment for import. That complicates the deployment of new services based on physical devices. But an open and disaggregated system can take advantage of hardware that has already been certified in the given market. That radically reduces the time it takes to turn up a new service in that locality – and it eases the depot requirements discussed above. That’s because the service is defined by the software application, not the hosting hardware. Each region could use a different brand of hardware, while the operator is providing the same retail service in every region.

Custom applications – In the early days of our NFV solution we were briefing the network engineers from an early telco customer. They had selected our solution for deployment of SD-WAN and firewall software. But they were intrigued by the underlying Linux operating system.

One asked, “Could we load our own maintenance application to run on each node?” Yes.

Another asked, “Can we run our custom monitoring program?” Of course.

At this point I said that the software was an open and standard hosting environment, and that they could run whatever they wanted. You could see the lightbulbs going off as the ideas came fast and furious. They selected the system to deliver a known application. But they could use it to deploy their own proprietary applications to simplify network operation and maintenance – independently from the supplier (us).

Innovation with disaggregation may seem scary, but we can help

At ADVA, we’ve long been supporting open and disaggregated systems across our product lines. That includes open line systems, virtualized services hosted on open compute, and software-define networking hosted on open switches. Let us help you discover the obvious AND hidden benefits of this approach.