Build for the cloud: Five steps to modern networks

Is your network ready for the future?

Not long ago, the definition of a cutting-edge IT network was a fleet of static desktop PCs with a robust corporate data center as its beating heart.

Today, that seems … almost quaint.

An entire continuum of cloud solutions is taking over, as countless laptops and mobile devices continuously connect to services like Microsoft's Office 365 and Salesforce.com or applications on AWS, Azure and Google Cloud Platform. A growing array of sensors from connected machinery and edge devices are churning out data while a new army of remote workers need flexible “anywhere, everywhere” access.

The trouble is, those old-school enterprise networks haven’t been able to keep up.

Many have hardly changed since those days of desktops and data centers. Instead, they’re trapped in a cycle of pasting unstable workarounds over an aging and outdated infrastructure that can’t handle the load.

In large organizations, for example, more than half of the network budget can be spent on bandwidth. And with the incessant demand for new cloud services, those bandwidth needs can grow 30% each year—with a similar impact on cost.

Why have networks fallen behind? The truth is, adapting to a cloud-based system is too often treated as an afterthought or as “just another” infrastructure fix. That inevitably leads to a patchwork approach, where issues are addressed only when something breaks. Organizations get stuck in a spiral of technology debt, where their network teams spend all their time and budget wrangling legacy technology rather than reimagining it for the business’s future needs.

Why overhauling your network should be a priority

Today’s networks need to be able to meet a wide, rapidly changing set of needs as well as provide employees with seamless connectivity to data, applications and platforms from anywhere and everywhere. This includes the ability to handle:

To meet those needs, you can’t rely on quick fixes—you need to rearchitect the network entirely for a modern environment.

That means building an any-to-any device-to-cloud network that leverages an advanced automation layer and is linked to a broader infrastructure-as-code approach. Here, we’ll walk you through practical ways to get your network from a position of technology debt to a position of technology wealth.

How enterprise networks can boost cloud continuum experience

Five steps to a modern network

1. Build a secure network foundation for cloud services
   
   The first step is to make sure your new network architecture forms a solid, safe foundation for your cloud strategy. That means designing in rigorous security to cover a complex web of traffic. Network servers that once lived in data centers should be migrated to the cloud for better integration. The emergence of 5G offers the chance to “remove the wires” of private networks to make enterprise-wide mobility of devices easier. By creating a configurable architecture, the network is ready to change as new cloud-native projects are added.
   
   Delivering global cloud architecture. Accenture helped one large manufacturer achieve this via its Secure Cloud Foundation framework—a solution that automates the setup of an environment rooted in a secure baseline that can support scalable workloads in the cloud.


2. Brace for a hybrid/multi-cloud environment
   
   No organization has all of its workloads in a single cloud—hybrid/multi-cloud is inevitable, and networks need to be designed to meet this need.
   
   Rather than treating networking as a problem to be solved (and re-solved) in each new cloud project, the goal is to provide flexibility so organizations can choose options that meet their unique needs—whether that’s machine learning services on the Google Cloud Platform, a data lake in Azure or workloads in AWS. To simplify this process, consider using a software layer product to manage the WAN and 5G connectivity across multiple clouds.
   
   Layered approach. Accenture used a similar strategy for a major pharmaceutical company that needed help building a single unified network service. Now, the company has a more cost-effective system with seamless user experience—and the system is flexible enough to scale and evolve along with the business strategy.


3. Automate to recreate
   
   Back in the day, networks were static environments, with routers and switches added as required and engineers to manage the physical hardware. But that pattern can’t hold up in a cloud-based world. A cloud service might spin up thousands of workloads instantly to handle a sudden burst of activity. Managing a network manually at that speed? Impossible.
   
   Instead, you’ll need a highly automated programmable network layer integrated with cloud workloads and enterprise tools via a set of APIs. That way, network monitoring and analysis are unified—vastly accelerating the network’s ability to flex with the needs of the business.
   
   Automation at work. At a large oil and gas company, Accenture took over operations of the client’s global network and deployed more than 125 pre-built use cases for different automation scenarios to quickly reduce operations team activities. Using a variety of tools, Accenture customized those automations into the production environment. As a result, the company saw an 80% reduction in incidents.

4. Resilience by design
   
   The best networks are those built with the idea that they’re going to fail.
   
   Think about it: Sending an engineer out to fix network hardware every time an outage happens can be costly. But if you build redundancy into the network from the start, you create resilience across the whole system. That way, no single outage can ever impact users’ abilities to access any services.
   
   One way to do this? Invite chaos. Under a “chaos engineering” model, developed initially by Netflix, a tool (referred to as a “chaos monkey”) causes outages at random to test the resiliency of the system. The enterprise can then address weaknesses, strengthen the overall system and have confidence it can deliver round-the-clock availability.
   
   Keep it simple. In lieu of controlled chaos, another strategy is to focus on simplification, as Accenture did with a large insurance company. After developing automations that could spot problems, log tickets and schedule fixes, the time it took to resolve outages dropped by 35%.


5. Build for new ways of working
   
   An organization’s goal should be to let everyone access cloud services from anywhere and everywhere—but with the right level of security, the right bandwidth, the right performance and at the right cost. Secure Access Service Edge (SASE) is an emerging option. By bundling together SD-WAN and security capabilities, delivering them as a service, SASE provides a safer, easier way to allow access the network from any location.
   
   Smarter security, everywhere. Accenture designed and deployed a cloud-based network architecture for a chemicals company that integrated security across all layers including SASE. This allowed services to be delivered securely for users from offices and remote locations. When the pandemic hit just a few months into the transformation, the company was ready: All users were able to seamlessly transition to remote work thanks to the integrated SASE architecture.

Take your network from tech debt to tech wealth

The way forward for networks is clear: Companies need to abandon short-term “patchwork” fixes and focus on fundamental network upgrades for the cloud. Those old network bottlenecks limit innovation, escalate costs and open potentially disastrous security holes.

With a new network approach, one that uses a programmable layer to deliver instant scalability, fast automation, more flexibility and better security, businesses will be able to stop throwing money and time at problems—and start seeing better results.

Related insights