Continuous Network Assessment Is The Key to Modern NetOps
Evolving network operations need continuous network assessment. Here’s why.
As hybrid networks have become more cloud-connected, expansive and complex, the typical NetOps managers’ device-level mindset isn’t sufficiently geared toward preventing outages and maintaining a strategic service delivery capability. We need a continual service-delivery assessment mentality, says Song Pang of NetBrain Technologies.
How do you know when an enterprise network is “working”? Ask a network operations (NetOps) manager, and they might say something about the status of all the various network infrastructure components being operational. And they may then go on to say that the network is “up” when all those devices are reachable. But this kind of device-oriented, binary response represents a dated way to think about network performance.
In a digital-centric world, network success should be defined by the ability of the infrastructure to deliver critical business services. An enterprise’s long list of design, compliance and application-specific use cases each has a specific set of needs that need to be viewed from the top down concurrently. Without such a service delivery view, it’s easy to lose track of what really matters. For instance, a network that works well for ecommerce might not work well for VoIP traffic, or a network that allows high-performance data access may also present too large of an attack surface.
At the end of the day, it’s the business that determines the long list of operational parameters of the network that must be maintained concurrently. These parameters include enforcement of specific types of security profiles, the ability to maintain specific levels of throughput for each of the applications deployed, and so on.
It turns out “working” has many shades of gray, and the context of the question is paramount. In today’s digitally connected business, a “working” network is one that allows all the applications and services needed by the business to be delivered successfully.
A More Strategic Approach to NetOps
In addition to updating their approach, it’s important for network operations managers to rethink the very processes that are needed to scale their support plan without the typical associated labor overhead, become more proactive in preventing outages, and ensure that the network-driven business risk is low. They need to view the network to deliver services rather than a collection of devices that support those services.
See More: Network Management: Why NetOps Is the New DevOps
Identifying goals
When trying to refocus their work through the lens of assessing service delivery, NetOps must first determine what services the network is trying to deliver. In other words, what are all the stakeholders’ needs or “intents” of the network? This often means maintaining certain levels of performance for specific applications, maintaining certain security and design compliance standards, and being resilient to problems through well-conceived failover architectures.
There is another level of granularity below each of these broad categories that affects service delivery. For example, an application such as voice-over-IP requires certain conditions to be met for calls to sound crisp. Whatever the architecture, NetOps must be constantly verifying through broad assessment that the network is delivering the needed services and capabilities.
Start with baselining
The first step is for NetOps to identify the expected network behaviors or baselines required for each application service. By articulating the desired behaviors as defined by all the operational groups across all the mission-critical applications, they can create a comprehensive list of behaviors for assessment.
Once the baseline expectations are well understood, it’s a simple matter of verifying or assessing those desired behaviors against the actual network in real-time. Previously, this step often got derailed when the network was too large, too distributed or lacked clear visibility into the various components and public clouds. A June 2023 Uptime Institute survey shows that half of all computing is now occurring in the public cloud, and 44% of respondents to another 2023 survey identified cloud visibility and network security as a top concern. Full assessment of all aspects of these hybrid environments is a necessity for successful service delivery.
Add some automation
No-code network automation is a requirement for this new mode of continuous assessment. Why? The short answer is scale. We always know how to assess any part of the network once. The most capable engineers even know how to manually determine service delivery capabilities. But that one-off approach is very engineer-dependent and consumes a great deal of time. Once you multiply the points of interest by tens or hundreds of thousands across a complex enterprise network and then assume a daily or weekly refresh, you see why scale is the issue. Not to mention the other types of assessment that spring into view once you have an automation platform available. These include security and compliance, application specifics, and cloud performance.
NetOps teams and budgets have not kept pace with this scale or complexity. As a result, it’s simply not feasible for NetOps or security teams to manually check the delivery of each of their required application services, verify the configurations of hundreds of routers, switches or firewalls, or test that all the attack surface for the infrastructure is well understood and uncompromised. Automating this kind of deep, continuous assessment across the entire network is both necessary and strategic.
Benefits of a Service Delivery Mindset
Focusing on service delivery (rather than connectivity and device health) allows NetOps to be more aligned with the actual needs of the business. While resolution to problems detected can take many forms, continuous assessment spots minor errors before they become major errors. This kind of assessment assures business leaders that their infrastructure is fully supporting their application services.
Change management is a great example of the importance of prioritizing service delivery rather than focusing on device health. In fact, the vast majority of network outages are caused by the unintended consequences of routine network changes like updates or security fixes. A connectivity-based outlook often misses these unexpected consequences since it only focuses on devices and their ‘normal’ operation. With no context of the business that traverses each device, the changes made to a given device can be deemed “successful,” only to realize through the service desk that a given change has impacted another business service.
See More: Achieving Digital Agility Through Change Management
Thinking Strategically
Too often, NetOps focuses on outcomes that are out of step with the business needs of the network. All too often, the result is a scenario where the operational tasks are executed successfully, but that change creates a litany of unintended consequences. A service delivery assessment approach eliminates the disconnect between problems, services, remediation and business goals. Continuous network assessment at scale dramatically simplifies and lowers the cost of network operations and keeps network operations tightly aligned with the needs of the business itself.
What new strategies have you introduced to network operations? Share with us on Facebook, X, and LinkedIn. We’d love to hear from you!
Image Source: Shutterstock
MORE ON NETOPS
