For decades, enterprise IT has operated with a profound operational disconnect: the teams managing applications and the teams managing the underlying network have largely worked from entirely separate universes. Application performance management platforms and configuration management databases track software details, code repositories, systems, and databases. Meanwhile, entirely separate network tools, cloud consoles, and microsegmentation systems monitor paths, cloud resources, and security controls.
This historic fragmentation has long limited what operations teams—and increasingly, automated AI agents—can reliably determine about the physical and virtual infrastructure supporting a given business-critical application.
IP Fabric is now aiming to fundamentally solve that visibility gap. The company, known for building read-only digital twins of enterprise network infrastructure to discover devices, paths, and configurations, has released version 8.1 of its platform. The latest update introduces application infrastructure mapping, connecting application workloads directly to the network and cloud paths they depend on, alongside a completely redesigned cloud-native data model. This release directly follows IP Fabric 7.9, which previously expanded discovery and path analysis capabilities for Azure and Google Cloud Platform environments.
"For the first time, we will now have the service-aware digital twin, meaning that you can ask how this application path looks like today," said Pavel Bykov, CEO of IP Fabric, in an interview with Network World.
Where Application Maps Meet the Network
The historical divide between application and network domains has created persistent hurdles for enterprise troubleshooting and change management. Application teams and network teams have spent generations looking at fundamentally different maps of the same digital ecosystem.
"For a really long time, like for decades, the world of applications, systems, databases, code, workloads, and the world of networking, the cables, the connections, the cloud, were separate," Bykov explained.
To bridge this divide, IP Fabric 8.1 treats applications, workloads, and flows as first-class objects directly within the platform. For every registered application, the system calculates the comprehensive end-to-end path for every underlying workload it depends on, successfully assembling those complex routes into a unified dependency map.
Crucially, this mapping capability works bidirectionally. Enterprise teams can query the system to discover precisely what infrastructure a given application depends on, or conversely, which applications might depend on a specific network device or link before it is scheduled for an upgrade or taken offline for maintenance.
Furthermore, the platform can automate the generation of daily dependency reports alongside pre- and post-change impact summaries. These detailed summaries can be seamlessly fed into external automation tools and emerging AI agents to streamline operational workflows.
Bykov pointed to a common, highly frustrating troubleshooting scenario to illustrate the practical value of the update. Imagine a scenario where a network link appears overloaded and traffic is dropping. When an engineer compares the current path against the previous day’s model, they might discover that the application did not merely suffer from congestion; it actually shifted to an entirely different path across the network.
While IP Fabric has always excelled at discovering native network infrastructure automatically, application data represents a fundamentally separate challenge that the platform does not discover directly. Instead, that critical context is ingested from external sources. The journey begins with a strategic partnership with Illumio, and Bykov noted that further native integrations with other security and management vendors are already planned. Additionally, teams can import application data flexibly through CSV files or direct API connections.
Modeling Cloud on Its Own Terms
Another major pillar of the IP Fabric 8.1 release involves a complete overhaul of how the platform approaches cloud environments. In previous iterations, IP Fabric captured cloud data and normalized it into a traditional data model built primarily around on-premises concepts, such as next hops and physical hardware devices.
According to Bykov, that traditional approach ultimately failed to hold up under the realities of modern cloud architectures, which include traffic manipulators, gateways, and transformers that possess no direct on-premises equivalent.
To resolve this limitation, IP Fabric 8.1 introduces a redesigned cloud-native data model where VPCs, VNets, subnets, virtual network interfaces, route tables, peerings, and security policies are treated as first-class entities rather than forced into an on-premises shape. Underneath the user interface, a translation layer still exists to normalize objects like transit gateways and load balancers, but it now translates them into a cloud-native model rather than an outdated on-premises worldview.
This architectural shift significantly enhances path analysis within a single cloud environment, across peered networks, and between cloud and on-premises infrastructure. The platform actively evaluates AWS Security Groups, network ACLs, and Azure Network Security Groups to determine with high precision whether a given traffic path is permitted or blocked, while instantly identifying which specific security policy is responsible for the outcome.
"We are working with how the cloud is actually communicating instead of trying to basically massage that model onto the on-prem worldview," Bykov said.
AI and the Path Forward
The introduction of application mapping and cloud-native modeling arrives at a critical time for enterprise artificial intelligence initiatives. IP Fabric already enables AI agents to query the platform via its hosted Model Context Protocol (MCP) server, a capability that predates the 8.1 update. However, the latest release dramatically deepens what sits behind that interface, allowing automated agents to query the same verified application and cloud context through the exact same server.
Bykov emphasized that the stakes for secure, accurate infrastructure access are rising rapidly as AI agents evolve past simple advisory roles. Increasingly, AI agents are logging directly into network devices and executing configuration changes independently, heightening the operational necessity of validating what those automated modifications actually achieved in practice.
"We all know that AI is only as good as the integrations to other systems, and there is now this one interface to the network," Bykov noted.
Looking beyond the deployment of version 8.1, Bykov outlined the company’s roadmap, pointing toward continued engineering efforts focused on scaling the platform to handle even larger and more complex enterprise networks, alongside expanding vendor and cloud coverage.
"General direction is definitely right now more system improvements to cover larger, more complex networks," Bykov concluded.