Bringing Cloud-Native Network Observability to Nutanix Flow with Retina, developed by Microsoft

Modern applications are increasingly built on Kubernetes®, supporting agility and scalability for enterprises embracing cloud-native architectures. But as organizations deploy more distributed services and containerized workloads, gaining visibility into application and network behavior becomes more challenging.

Customers often tell us they want a consistent operational approach for Kubernetes environments alongside their traditional virtualized infrastructure. They need the ability to understand network performance, quickly troubleshoot issues, and maintain application reliability without introducing complexity or performance overhead.

We're excited to share the next step in the ongoing collaboration between Nutanix and Microsoft: integration between Nutanix Flow, a dual-native networking and security platform for Kubernetes and virtual machine environments, and Retina, developed by Microsoft, an open-source observability platform built on eBPF technology. Together, Flow and Retina combine policy, security, networking, and deep observability into a unified solution, providing Kubernetes operators with end-to-end visibility, faster troubleshooting, and simpler operations while preserving the enterprise-grade experience customers expect from Nutanix.

Expanding the Nutanix and Microsoft Partnership

Nutanix and Microsoft have a long history of working together to deliver solutions that help customers modernize infrastructure and applications. 

Microsoft introduced Retina as an open-source, cloud-native Kubernetes network-observability platform in 2024, designed to provide deep visibility into Kubernetes networking environments. Retina leverages eBPF (Extended Berkeley Packet Filter), a modern Linux kernel technology that enables efficient collection of networking telemetry without requiring intrusive modifications to applications or workloads.

By integrating Retina capabilities with Nutanix Flow, we’re enabling customers to gain enhanced visibility into Kubernetes networking environments running on Nutanix infrastructure.

The collaboration also supports ongoing work to explore additional observability capabilities and advanced solutions for enterprise Kubernetes environments, subject to future product plans and availability.

Building on Nutanix Flow for Containers

In December 2025, Nutanix expanded the Flow portfolio with Flow CNI, extending Flow capabilities into Kubernetes environments through a Container Networking Interface (CNI) built on Open vSwitch (OVS) and Open Virtual Network (OVN)-Kubernetes.

Unlike standalone Kubernetes networking solutions, Nutanix Flow provides a dual-native approach spanning both virtual machine and Kubernetes environments. By combining Flow's unified networking and security capabilities with Retina's deep observability, customers gain consistent visibility and operations across both traditional and cloud-native applications.

Flow CNI was designed to provide networking and security for Kubernetes environments while extending Nutanix's dual-native operational model across both containerized and virtualized workloads, all commercially backed by Nutanix's 90+ NPS support.

As Kubernetes adoption accelerates, however, networking teams may also need more than connectivity and policy enforcement. They also need actionable insights into how applications communicate, where problems occur, and how network behavior impacts application performance.

"Our goal with Nutanix Flow is to provide simple, enterprise-grade networking and security for cloud-native as well as virtualized workloads," said Saveen Pakala, VP of Product Management at Nutanix. "Integrating Retina's eBPF-based observability gives our customers deep, granular visibility into Kubernetes traffic without introducing complex overhead, empowering operations teams to ensure high availability and rapid troubleshooting."

Why eBPF-Based Observability Matters

Traditional monitoring approaches often rely on packet mirroring, sidecars, application instrumentation, or multiple disconnected tooling frameworks. While valuable, these approaches can introduce operational complexity, consume resources, or provide only partial visibility.

Using eBPF technology, Retina can collect networking telemetry from the Linux kernel, enabling highly granular visibility into Kubernetes traffic while minimizing impact on application performance.

Because Retina is CNI-agnostic, it works across Kubernetes networking environments while providing a consistent observability framework.

For organizations running Kubernetes on Nutanix, this creates a powerful combination:

  • Commercially supported, enterprise networking with Flow CNI
  • Deep Kubernetes network observability, powered by Retina, developed by Microsoft 
  • Simplified operations through integrated visibility and insights

"Retina was created to deliver deep, cloud-native network insight using eBPF technology," said Chandan Aggarwal, Partner/General Manager at Microsoft. "By collaborating with Nutanix to integrate Retina with Nutanix Flow, we are helping enterprises gain comprehensive network visibility and streamlined operations across their Kubernetes deployments."

Deep Visibility Across the Kubernetes Stack

The integration of Retina with Nutanix Flow is designed to provide visibility across multiple layers of the Kubernetes environment, helping platform teams understand network behavior, accelerate troubleshooting, and maintain application performance.

Key areas of visibility include:

  • Infrastructure visibility to monitor network throughput, packet drops, connection counts, resource utilization, and other node-level metrics that help identify congestion, capacity constraints, and emerging infrastructure issues.
  • Service communication visibility to understand how applications interact across the cluster, including pod-to-pod communications, service dependencies, traffic flow paths, and east-west traffic patterns that can impact application performance.
  • DNS and connectivity insights that reveal DNS queries, resolution failures, TCP connectivity data, and flow-level networking information to help operators quickly isolate application communication issues.
  • Packet-level and network health analysis that provides visibility into TCP resets, API server latency, connectivity anomalies, and on-demand packet capture capabilities for advanced troubleshooting and root-cause analysis.

Together, these capabilities provide a comprehensive view of Kubernetes networking, enabling teams to move from reactive troubleshooting to proactive operations.

Reducing Mean Time to Resolution

For many organizations, observability is not simply about collecting more telemetry. It's about accelerating operational outcomes.

When application issues occur, teams need to answer questions quickly:

  • Is the issue occurring at the network layer?
  • Which service is experiencing the problem?
  • Is traffic reaching the intended destination?
  • Are DNS failures contributing to application errors?
  • Is latency increasing between specific services?

By combining Flow CNI networking with Retina observability capabilities, customers can move from reactive troubleshooting to proactive operations.

The result is fast issue identification, efficient operations, and greater confidence when running production Kubernetes workloads.

A Shared Vision for Cloud-Native Operations

Kubernetes environments continue to grow in scale, complexity, and business importance. As a result, observability has become a foundational requirement rather than an optional capability.

The integration of Retina with Nutanix Flow reflects a shared commitment between Nutanix and Microsoft to simplify cloud-native operations while providing the enterprise-grade visibility customers need.

Together, Nutanix and Microsoft are helping organizations gain deeper insight into Kubernetes environments. By combining Nutanix Flow's dual-native networking and security capabilities with Retina's eBPF-powered observability, customers can apply a modern, consistent operational model across both traditional virtual machines and modern cloud-native applications.

Learn More

To learn more about Retina, developed by Microsoft  and the open-source project, visit:

As we continue to collaborate with Microsoft, we're excited about the opportunities to deliver even deeper observability and operational insights for cloud-native applications running on Nutanix.

©2026 Nutanix, Inc. All rights reserved. Nutanix, the Nutanix logo and all Nutanix product and service names mentioned are registered trademarks or trademarks of Nutanix, Inc. in the United States and other countries. Kubernetes is a registered trademark of The Linux Foundation in the United States and other countries. All other brand names mentioned are for identification purposes only and may be the trademarks of their respective holder(s).