The Role of Automation in SD-WAN Optimization

A recent survey of IT and networking professionals shows that almost all of them (97%) believe that using artificial intelligence (AI) and machine learning (ML) to drive automation in software-defined wide area network (SD-WAN) environments is essential. Some even view it as critical. The research highlights how AI is expected to enhance automation and boost operational efficiency in complex SD-WAN solutions.

The survey, conducted by TechTarget’s Enterprise Strategy Group (ESG), gathered responses from 374 participants involved with networking technology in the US and Canada. It indicates that SD-WAN environments need to evolve and become more dynamic as IT infrastructures grow more distributed and complicated.

Network operations teams recognize the need to be more proactive, improving their ability to detect issues faster (mean time to detect or MTTD) and fix them quickly (mean time to repair or MTTR). AI, ML, and automation are expected to help achieve this. Of those surveyed, 40% identified detecting unusual activity as important, 39% mentioned predictive analytics to catch issues early, and another 39% cited faster troubleshooting as key benefits of AI in their SD-WAN solutions.

AI will also be used for making recommendations, optimizing performance, and potentially automating fixes without human intervention once it’s fully trusted. Given the increased risk from a broader attack surface, ESG analysts say it’s encouraging that businesses plan to use SD-WAN to speed up issue detection and response.

AI Features in Network Equipment

Recognizing the benefits AI offers for network operations, equipment providers are adding AI and ML to their product lines, enhancing AIOps (AI for IT operations) to support network functions. This innovation is expected to further boost SD-WAN solutions and SD-WAN managed services.

In February 2024, Cisco introduced AI-enhanced networking, security, and visibility tools at Cisco Live 2024. These tools are designed to give businesses insights to secure their digital footprint and tackle core challenges. Similarly, in April, Extreme Networks launched AI Expert, a service that collects data from across the network to boost performance and efficiency. AI Expert provides insights, automates operations, and alerts users to network issues like overload or Wi-Fi dead spots, enhancing their SD-WAN managed services offerings.

Juniper Networks followed in June with new AI tools for SD-WAN solutions, including the extension of their digital assistant, Marvis Minis, to SD-WAN operations. This AI tool diagnoses authentication issues, runs continuous speed tests, and helps resolve network problems before they escalate, all without human intervention.

Generative AI Helping Network Admins

Generative AI (GenAI) is emerging as another key tool in networking, especially with the growing IT skills shortage. According to John Burke, chief technology officer at Nemertes Research, many new IT professionals focus on general IT skills rather than specialized networking expertise. GenAI could assist less experienced IT staff in managing complex SD-WAN networks by automating routine tasks and responding to incidents.

Burke believes GenAI will mature to the point where it can act as a “network administrator copilot,” helping teams document SD-WAN solutions, generate policies, and even audit configurations. GenAI could also assist with network scripting, though Burke cautions that network engineers should carefully review and tweak any code the AI generates.

Looking Ahead: AI in SD-WANs

The addition of AI features to network management tools reflects the industry’s recognition of the growing complexity of corporate networks. As these networks become more complicated, the demand for network professionals will increase. While fully automated network management isn’t on the horizon, AI tools can provide valuable support, especially for businesses utilizing SD-WAN managed services and SD-WAN solutions, making network operations more manageable.