How to Get the Maximum Value Out of Network Automation

Network automation takes advantage of programmable logic to manage network services and resources. The approach allows network operations (NetOps) teams to configure, scale, protect, and integrate network infrastructure and application services far more rapidly than when performed manually.

Until relatively recently, most network operations were handled manually. "With AI coming, everything needs to be faster, better, and simpler, so you can't continue to be so manual," says Mattias Fridström, vice president and chief evangelist with Swedish telecommunication services provider Arelion, in an email interview. "Operators who understand this, and get their network automation strategy under control in the coming years, will position themselves for success."

Addressing challenges with network automation

Nikos Katinakis, CTO at telecommunications infrastructure company Zayo, observes that most networks have different buckets of automation. To get the maximum value from network automation, he recommends building a plan that addresses each of the buckets, ensuring they all come together at some point. "It's a complex puzzle that needs a strategy to solve," he says via email. Point solutions, patching one small part of the problem, can provide immediate relief at a specific pain point, yet they don't really move the organization forward in any meaningful way. Katinakis suggests that automation, AI, and analytics should be used as strategic tools to address short-term efficiencies as well as long-term impacts, such as improving customer experience and guiding decision-making for future business growth and innovation.

For multi-cloud adopters, an immediate challenge is ensuring that all involved parties possess the same understanding of multi-cloud network management, says Sandeep Vohra, head of strategy, Americas, at global communications integrator Orange Business. "There are distinct types of multi-clouds, each with its own unique management requirements," he notes in an email interview. Public cloud-to-public cloud is one type; another is private cloud-to-hyperscaler cloud. "With so many different applications and platforms in play, all parties must work within the same definition of multi-cloud," he notes. Without it, stakeholders are adding extra fragmentation on top of the existing technical complexities.

Major dangers

Security and data sovereignty are frequently major blind spots in multi-cloud management strategies since they involve complex and evolving regulatory requirements, as well as a wide range of potential security risks and threats, Vohra says. When organizations use multiple cloud providers and platforms, they must ensure that their data is protected and compliant with various regulations, such as GDPR, HIPAA, and PCI DSS. "This can be challenging because different cloud providers may have different security and compliance standards, which can lead to inconsistencies and gaps in security policies and controls."

Typical multi-cloud vulnerabilities include unauthorized access, data breaches, and cyber-attacks. "These risks can arise from various sources, such as misconfigured cloud resources, insecure APIs, and unpatched software vulnerabilities," Vohra says. "Furthermore, the complexity of multi-cloud networks can make it difficult to detect and respond to security incidents in a timely and effective manner."

Seeking value via network automation

The best way to achieve maximum network automation value is to focus on the organization's own data model, which should cover all current systems and applications, Fridström says. "Ensure you know where your master data is stored so you have one single source of truth."

Fridström adds that effective network automation value often comes when combining data sources to find anomalies or similarities. "This approach includes having a correct and connected inventory system, which supports automation in many processes," he explains. "Having your data under control also allows you to share more data with customers so they can use that data for self-service purposes."

Many organizations unintentionally hamper network automation by adding new data storage systems that are siloed, Fridström says. "Initially, you may feel that an equipment vendor's management system is a great place to store some of your network inventory data," he says. "But if that system isn't connected and synchronized properly with other similar systems, you'll soon have two sources of truth."

Katinakis stresses the importance of internal collaboration to lift automation to its full potential. "As CTO, much of the responsibility lies with me in driving and implementing the automation plan," he states. Still, the final plan should be an aligned effort across the entire organization. "When it comes to developing and deploying new and innovative technology, all teams—engineers, security experts, sales, product managers—must be aligned on the vision, and be willing to change the way they interact with the network and the way they work."

Looking forward

"We are at a technology inflection point where the emergence of accessible AI allows us to rethink how we work," Katinakis says. "In the network specifically, we’re able to automate workflows, provide recommendations, and make decision-making more efficient."

Related articles: