As our industry speeds towards a multicloud future, I feel like I have a somewhat unique perspective having spent a good part of my career on the user side architecting, building, and operating large-scale cloud infrastructure at Google. It’s one thing to deliver products that help solve problems, but it’s another thing to be the one responsible for actually solving the problems.
If I had to reduce everything I have learned from being a customer to a single point, it’s this: wake up every day with the primary objective of making things simpler.
The future belongs to the fast.
In our industry speed is a matter of survival. When you consider how fast competition is changing, how fast technology is changing, how fast consumption is changing - clinging to TCO as a security umbrella is a losing proposition.
Agility shows up in different ways depending on the business. If you’re in an industry that is witnessing disruptions in business models every day, being fast is a matter of getting new products and services to market. The payment industry, for example, has seen a marked increase in velocity as company’s flock to the space using technology as a lever to drive improved loyalty programs, anywhere banking, and simplified transaction handling.
It could be that agility is related to security. You need to be able to quickly react to the latest vulnerability or malware outbreak, and that requires having the tools and processes to move fast when you need to.
Or maybe it’s agility in a more general sense, and all you really want to do is take advantage of the latest technology. Consider that every year for the last seven years, networking has introduced a new paradigm-shifting breakthrough: white box, SDN, NFV, overlays, DevOps, and now intent-based networking.
Whatever the driver, it’s clear that being fast matters.
Complexity is an anchor.
So, what keeps us from being fast? It’s complexity. The truth is that we have made networks very complex to manage. Years of incrementalist thinking mean that we are always adding on — and that stopped adding up a while ago.
It’s not just that making changes is complex. Rather, it’s that the layers of complexity have led to a culture that actually fears change. Most enterprises are downright draconian about changing infrastructure, especially during critical times. Why? Because we are too fragile and change threatens to topple the house of cards.
And when you combine a fear of moving with a need to be fast, you introduce a level of risk to the business that is simply intolerable.
Enterprises are the most complex environments that exist.
A lot of people look at cloud companies with the view that they run the most complex environments on the planet. It’s just not true. They are certainly operationally advanced, but that’s because they have the advantage of homogeneity, allowing them to optimize operations in ways the average enterprise simply cannot.
Complexity is a function of a number of different things: the number of connected devices, the diversity of software running on those devices, the number of applications, the number of users – the list goes on. If you look at the hyperscalers, they have spent years creating highly homogeneous operating environments. The devices look similar, the software is uniform and the configurations are largely identical.
Enterprises don’t have the luxury of declaring everything end-of-life. They have to support different lines of business, legacy applications and distributed teams with unique requirements. This means complexity is going to exist. The real question is who handles that complexity: the user or the vendor?
These two words, more than any other reason, are why I came to Juniper Networks. To make things simple, you can’t just dumb them down. We recognize that you don’t make things complex for complexity’s sake. The organizations are actually complex. So, simply declaring that things should be simple and ignoring the realities of a business is a non-starter. The path to simple runs through complexity. You have to address the complexity and that requires real engineering strength. You get to simplicity only through superior engineering.
Making it all real.
If multicloud is the destination, then we come to terms with the fact that we are entering an even more complex operating environment – a patchwork of domains across which workloads need to be distributed and orchestrated.
The answer for handling edge policy and overlays has historically been to build an overlay that has very little resemblance to the actual underlay and then to manage each independently. If complexity is the enemy, turning one management domain into two might simplify policy and control, but it does so at the expense of increased complexity.
We believe that the underlay and overlay should to be controlled as one. And that is why we have enhanced our Contrail controller for the enterprise as Contrail Enterprise Multicloud. We have essentially made it possible for customers to manage their overlays, underlays, public cloud workloads and distributed firewalls (as part of a microsegmentation solution) all from a single platform.
In essence, we have taken the complexity inherent in a multicloud environment and made it our problem to solve, leaving our customers a much-simplified experience. This is the power of engineering simplicity.