As part of APMdigest's 2025 Predictions Series, industry experts offer predictions on how NetOps, Network Performance Management, Network Observability and related technologies will evolve and impact business in 2025.
AI DRIVES NETWORK TRANSFORMATION
Enterprise AI initiatives will drive network transformation. AI workloads in data centers will require network teams to invest in new Ethernet fabrics. They will also need to update their wide-area networks as AI training and inference will require significant uploads of data from the network edge. EMA expects that NetOps teams will need to upgrade their observability solutions, too. They will need real-time visibility into packet loss and latency. They'll also need tools that offer predictive analytics around spikes and bursts in AI-related traffic to ensure that the network can adapt quickly to volatile AI traffic requirements.
Shamus McGillicuddy
VP of Research, Networking, Enterprise Management Associates (EMA)
Listen to the MEAN TIME TO INSIGHT Podcast Episode 10: Generative AI
Click here for a direct MP3 download of Episode 10
AI NETWORKING
In 2025, enterprises will begin to make strides towards more intelligent networks, driven by advancements in the availability of AI. As global data volumes continue to rise, organizations will start to explore moving from reactive network management to proactive and predictive self-operating networks. However, traditional network management approaches may still be prevalent as enterprises gradually deploy AI applications that demand low-latency, high-bandwidth connections for real-time processing. Those organizations that do not begin the process of evolving their network infrastructure may encounter bottlenecks, impacting their ability to fully leverage AI technologies over time.
Jeff Gray
CEO and Co-Founder, Gluware
AI networking becomes useful in 2025. Networks have struggled to get AI models to understand network/telco data to drive business value, but that changes next year. We'll see network teams get both sides to communicate and open up the potential of use cases for AI and network practitioners, especially when it comes to replacing GUI. Just like with AI initiatives, we'll see immediate ROI, but fully expect AI networking to hit full stride in 2026.
Andrew Coward
GM, IBM Software Defined Networking
ENHANCED NETWORK OBSERVABILITY
Convergence of operational technology (factory automation systems, medical imaging, etc.) on IT networks will drive a need for enhanced network observability and security. Observability tool vendors will adjust their reporting and customization capabilities to accommodate new views of the network for OT operations. They will also need to collect data from newer network technologies like private 5G networks. There will also be an increased focus on zero trust principles that eliminate lateral network movement in OT environments, which will require zero trust microsegmentation. These new segmentation schemes will require enhancements to observability to help with designing, monitoring and enforcing these new security measure.
Shamus McGillicuddy
VP of Research, Networking, Enterprise Management Associates (EMA)
NPM SHIFTS FROM REACTIVE TO PROACTIVE
Network Performance Management (NPM) is shifting from a reactive approach to a more proactive focus, leveraging insights from network data to drive preemptive measures. The adoption of advanced network management protocols, like gRPC/gNMI, will gain momentum due to their efficiency in data transmission and seamless integration with data pipelines. Enhanced data extraction and integration with additional data sources opens up new opportunities for machine learning and closed-loop automation, enabling more effective and comprehensive network management.
Dan Wade
Practice Lead, Network & Infrastructure Automation, BlueAlly
CONVERGENCE OF NETOPS AND DEVOPS
As cloud adoption accelerates in 2025, it will be primarily driven by the need for scalable, secure, and cost-efficient solutions that enhance agility. As a result, Network Operations (NetOps) and DevOps will increasingly converge, adapting to a landscape where distributed, multi-cloud environments are the new norm. The shift to internet-based corporate networks demands not only robust network automation but a DevOps-led integration of observability and security across every layer. For example, by embedding AI-driven threat detection into continuous integration and deployment (CI/CD) pipelines, DevOps and NetOps teams can identify and address vulnerabilities in real time as new code is deployed. This close collaboration will be crucial for harmonizing network performance and security in a dynamic, cloud-driven environment, empowering organizations to safeguard assets and streamline connectivity in an increasingly decentralized world.
Erez Tadmor
Field CTO, Tufin
AUTOMATED NETWORK MANAGEMENT
We are seeing an evolution in the market toward consolidation of rampant tool sprawl, which has complicated IT troubleshooting. IT needs greater tool consolidation to reduce risk. Instances of shadow IT create concerns about people taking core IP code and running it through AI tools, posing huge business risks from a compliance and security perspective. Automated network management can drive compliance by allowing IT to know where all their systems and SaaS services are located. Armed with this knowledge, IT teams can create policies to lock down intellectual property and protect customer data.
Doug Murray
CEO, Auvik
CONVERGENCE OF NETWORK AUTOMATION AND SECURITY
Security and network automation will become inseparable in 2025, with integrated security validation becoming a mandatory component of any network change. Organizations will no longer treat network automation and security as separate disciplines but as two sides of the same coin. The rise in sophisticated cyber threats will drive this convergence, making automated security compliance checks and real-time vulnerability assessments and remediation standard features in network automation platforms. This will lead to a new category that I believe will redefine how we approach network security and automation.
Jeff Gray
CEO and Co-Founder, Gluware
INTENT-BASED NETWORKING PLATFORMS
In 2025, enterprises will increasingly recognize the necessity of advancing their network automation strategies, with a growing number of Global 2000 companies beginning to explore intent-based networking platforms. The convergence of traditional networking with more modern approaches will force organizations to rethink their automation strategies. Network teams will need to move beyond basic scripting and point solutions to embrace comprehensive, intelligent automation platforms that can handle multi-vendor, multi-domain environments. This transition will be driven by the increasing complexity of hybrid networks and the critical need to maintain security and compliance at scale.
Jeff Gray
CEO and Co-Founder, Gluware
IMPROVED USER EXPERIENCE
Network administrators will see more intuitive interfaces and personalized dashboards, thanks to AI-driven insights. So, while networks become more intricate, the user experience will become more minimalist, streamlined, and efficient.
Carla Guzzetti
SVP, Product Marketing & Customer Experience, US, Extreme Networks
NETWORK-AS-A-SERVICE
2025 will be a turning point as enterprises shift from traditional static networks to agile, on-demand, software-driven solutions. As enterprise environments grow more complex, managing connectivity across core, cloud, and edge locations has become increasingly challenging. Diverse locations, rising costs, and performance hurdles demand a fresh approach to networking. Enterprises need a seamless, end-to-end self-service experience, where automated network provisioning and real-time observability simplify operations and accelerate outcomes. In response, Zayo predicts a sharp rise in Network-as-a-Service (NaaS) adoption across bandwidth-intensive industries like healthcare, finance, and technology. By embracing NaaS, enterprises can unlock faster innovation, improved operational efficiency, and a future-ready network built to support AI and multi-cloud growth.
Bill Long
Chief Product & Strategy Officer, Zayo
SASE OBSERVABILITY
NetOps teams will need to address the blackhole of visibility introduced by secure access service edge (SASE) solutions. SASE vendors deliver security capabilities via cloud proxies or points of presence (PoPs) that often live in hyperscaler and colo environments. NetOps leaders are finding that they need better insight into how these proxies impact network performance and user experience. SASE proxies can add latency if traffic is routed inefficiently. These proxies can also experience performance and availability issues, like any cloud-based service. Today, most SASE providers offer little to know visibility into these proxies. EMA believes that SASE vendors will begin to offer granular reporting on SASE performance and start to make data available for export to third-party observability tools.
Shamus McGillicuddy
VP of Research, Networking, Enterprise Management Associates (EMA)
Listen to the MEAN TIME TO INSIGHT Podcast Episode 11: Secure Access Service Edge (SASE)
Click here for a direct MP3 download of Episode 11
SASE INTEGRATES AI
SASE will get more secure with AI. In 2024 AI agents demonstrated such groundbreaking vulnerability scanning and detection that they are expected to catch more than 25% of all new vulnerabilities in 2025. With this in mind, SASE frameworks may begin to integrate advanced AI models that proactively learn from real-time network behavior and threats. This evolution will allow the system to dynamically adapt security policies in milliseconds. Integrated AI allows SASE to go beyond traditional security controls by introducing context aware, predictive security measures that continuously fine-tune access rules, detect anomalies, and respond autonomously.
Prakash Mana
CEO, Cloudbrink
PERSONAL SASE
SASE gets personal. In 2025 companies looking for faster, simpler, more secure remote connectivity for their work from anywhere (WFA) employees will look beyond 1st generation Secure Access Service Edge (SASE) to personal SASE. SASE puts security and networking in the cloud for more worker flexibility. But most SASE architectures are still hardware based, focused on centralized locations, creating hairpins in the cloud, and don't account for users that might be connected to unreliable Wi-Fi or consumer broadband. Personal SASE shifts networking and security stack all the way to the user edge, lowering latency and increasing performance while still maintaining security.
Prakash Mana
CEO, Cloudbrink
NETOPS VENDOR CONSOLIDATION
NetOps tool vendor mergers and acquisitions will pick up in 2025. Large IT vendors and private equity firms will accelerate their acquisition activity around late-stage startups that offer network observability and network automation tools. As this occurs, a wave of new startups that leverage AI/ML technology for observability and automation will emerge and challenge incumbent vendors.
Shamus McGillicuddy
VP of Research, Networking, Enterprise Management Associates (EMA)
The network industry in 2025 will be a dynamic landscape shaped by consolidation, AI, and design. Expect fewer, but more powerful players as industry giants continue to merge.
Carla Guzzetti
SVP, Product Marketing & Customer Experience, US, Extreme Networks
Check back after the Holidays for more 2025 Predictions about Cloud/FinOps, DataOps and AI/GenAI.
