To help you and your team prepare for what the new year will bring to the network management space, I spoke with OpenText experts—Network Operations Management (NOM) product manager Pete Zwetkof and Network Automation (NA) product manager John Jackson. I wanted to get their perspectives on the trends that will be most relevant to network professionals in 2024. From the shift toward the cloud to advancements in AI and machine learning, here’s what they said.

Cloud workload shift: Enhancing network monitoring through cloud integration

How does the pivot of workloads to the cloud impact network monitoring strategies?

With more and more of the workload moving into the cloud, a good monitoring solution is critical. When we talk to our customers, we find that the same people who had been responsible for performance and fault resolution in the data center also have responsibility (especially for hybrid cloud applications) for making sure that endpoints in the cloud are reachable.

Our strategy is to extend management into the cloud by automatically discovering cloud assets like Virtual Private Clouds (VPCs), load balancers, rounding tables, firewalls, VPNs, and gateways—and monitoring them for performance and fault issues.

We’re also hearing from customers that they need to be able to monitor policy in the cloud. For instance, you may have a certain policy that says, “For these types of applications, I’m always going to use the transit gateway as opposed to a VPN.” You need to be able to understand your cloud-based applications because they change all the time. They’re inherently dynamic.

AI and machine learning: Driving efficiency in network automation and predictive maintenance

In what ways do you predict AI and machine learning will be used to advance network automation, particularly in areas like self-healing networks and predictive maintenance?

In 2024, AI can help users find and fix problems faster and also consolidate anomalies that are occurring on the network so they have a more complete picture. These capabilities could help drive higher availability. Self-healing is something that’s been built into networks for a long time. We use self-healing protocols like OSPF or BGP to allow us to work around problems in the environment.

With regard to what we’re doing here at OpenText, we have a solution that combines NOM with our Operations Analytics (Ops Analytics) product. We pass metric information and symptoms into Ops Analytics, where we then use machine learning to understand how metrics are related. When you’re trying to troubleshoot a complex problem that impacts critical business applications, understanding these metrics is essential. It brings you true network observability.

Adapting network automation for complex hybrid and multicloud infrastructures

With the increasing shift to cloud environments, how do you envision network automation adapting to manage hybrid and multicloud infrastructures effectively?

With the integration of SDN, SD-WAN, and SD access networks, networks are getting more complex. As a result, operators might be using multiple tools to manage their networks, which makes it tricky to maintain a comprehensive understanding of what’s going on. Half of the time, your customers are identifying problems before your operations staff, and that’s never a good thing.

We want to help our customers avoid that. Our strategy is to provide a single pane of glass, a single tool, to help manage all aspects of the network. A challenge, though, is the use of controllers (usually Linux boxes) to control SDx network elements. That’s why one of the things we’ve done is pivot to using REST for discovery, monitoring, change, and configuration. REST, combined with our adoption of new notification mechanisms like webhooks and telemetry, and our continued focus on vendor support, allows us to be that single pane of glass.

Overcoming the hurdles in network automation: Strategies for the upcoming year

What are the biggest challenges currently facing network automation, and how do you foresee them being addressed in 2024?

This was a big topic at the Network Automation Forum AutoCon 0 event I recently attended. One issue, like Pete mentioned, is that operators are using so many different tools. Another is that they could be using custom scripts that are only understood by the person who wrote them. If that person leaves the organization for whatever reason, processes break down.

In terms of a solution, the main idea here is to not boil the ocean. Start small, with low-hanging fruit—things that can be automated on a more granular basis. Then work up to the bigger picture.

There were a lot of other challenges that came up, too, like whether or not network engineers needed to develop coding skills, the feasibility of a single source of network truth, and integration with other tools and systems. With respect to network engineers developing coding skills, the consensus was that, yes, they’ll probably need to learn some level of programming (like Python or Go) to adapt to evolving NetDevOps environments.

In terms of working toward a single source of truth and integration with other tools, OpenText is well positioned to meet these challenges with the OPTIC Data Lake. The reality is that there will always be multiple sources of truth, but the OPTIC Data Lake is a critical source that can also help with that integration. You can interface with it via direct access to the schema or REST API, pull data into your own business intelligence tools, connect it to our UCMDB or Asset Management X products, or use the open data ingestion to push data from other outside sources into the Data Lake. A lot of these capabilities work out of the box and can be used very quickly.

Elevating user experience: The future of network monitoring in user-centric assessments

Given the growing emphasis on user experience, how do you think network monitoring will evolve to better assess and optimize user experiences?

There are a number of dimensions where UIs need to evolve. We need to focus on providing real insight into what’s going on in the environment, not just on providing data. The new UI experience we’re working on is centered on showing you the things that are most relevant in the moment. We want to take all of the data we have below the covers and really surface it in a way that’s useful.

One example would be our performance maps where we take performance data and make it relevant to topology. That topology context for performance problems, combined with our layer-three path views, aids in faster problem identification and resolution by transforming data into actionable insights.

Thanks John and Pete. Any other predictions for 2024?

Yes! I expect all the ex-Cisco Prime customers to be in a panic until they discover the perfect replacement and upgrade: OpenText Network Operations Management.

Check out our Practitioner Portal to learn more about:

OpenText Network Operations Manager (NOM)
OpenText Network Node Manager i (NNMi)
OpenText Network Automation (NA)