Arista Networks Inc. today released a new network observability software offering that brings together network infrastructure performance and data from compute and server systems of record to provide application and workload performance insights for data centers, campuses and wide-area networks.

Dubbed CloudVision Universal Network Observability, or CV UNO, the new software is aimed at reducing enterprise risk, improinge fault detection and correction, and simplifying cross-functional coordination using systems-level capabilities. The company says this approach speeds up time to recovery regardless of the root cause.

Historically, network engineers worked with management tools that functioned on specific domains. The campus network had one console, Wi-Fi another, data center a third and so on. This created the concept of “swivel chair” management, where a network engineer moved among multiple consoles and correlated the information manually.

Although this was never ideal, it worked when networks were static. Today, networks are constantly changing, and the volume of data networks generate has exploded, making legacy operations no longer viable. Arista’s approach is to deliver a single observability offering across the end-to-end network.

Arista CV UNO includes the following capabilities:

• Workload application and infrastructure discovery: According to Arista, CV UNO automatically identifies applications, hosts and workloads across diverse platforms and information technology systems. It offers a comprehensive view of the network and application landscape by combining data with CloudVision’s insights under Arista Network Data Lake, or NetDL. It also constructs an application-to-network graph that continuously updates, providing a historical perspective on the environment’s evolution. Arista’s NetDL is a single data source for the entire network and enables the company to map an application to a network. Without it, network professionals would need to stitch together information from separate systems manually.
• Proactive risk analysis: CV UNO facilitates proactive risk analysis through real-time application-to-network graphing within the change management process. It aids in cross-referencing and analyzing network issues and anomalies, enabling the assessment of potential disruptions before implementing changes in production and mission-critical networks. This information can be used to bridge the gap between security and networking teams.
• Real-time network change impact analysis: CV UNO analyzes and applies machine learning to the dataset within NetDL, identifying how network provisioning or state changes affect business and critical applications. It automatically pinpoints the impact of network alterations on application performance, empowering network engineering and operations teams to address issues swiftly. Tracking errors created from network changes can be difficult, if possible, with legacy tools. AI-based automation for change analysis is an ideal use case to introduce AI into the network.
• Host or application change impact analysis: CV UNO guides operators or engineers to the precise root cause of application performance issues, even when the network remains unchanged and without deploying host-based agents. This capability streamlines issue resolution and cross-functional coordination for the operations team.
• Topology-aware determination: The tool accurately discerns the root cause of application performance issues by consolidating insights from infrastructure systems, virtualization machines, systems of record and network flow and state data. This approach minimizes the tendency for blame-shifting associated with traditional methods.

Standard CV UNO components

• CV UNO Sensor gathers, standardizes, and organizes flow/SNMP data from various sources, such as VMware vCenter, DANZ Monitoring Fabric and third-party network devices, and transmits to NetDL.
• CloudVision-UNO is accessible via a premium feature license and integrates with and enhances CloudVision’s operational and network telemetry capabilities. It employs Machine Intelligence-based Analysis on NetDL-stored data to deduce topology-aware correlations among events, changes and anomalies, thereby speeding up root cause analysis and issue resolution.

CV UNO options

• The Recorder Node supplements packet capture, query, and replay functionalities to support intrusion detection, incident response and forensic use cases.
• The Service Node provides additional advanced packet processing functions, including end-to-end application latency analysis and DPI-based Application Identification and classification.
• The CV UNO Analytics Node offers distributed context-aware traffic analysis and machine-learning capabilities for large-scale optimization.

CV UNO will be generally available in the second quarter.

Some final thoughts

Arista is bringing together multiple network domains with full application visibility and troubleshooting, which will streamline network operations and improve uptime and reliability. With disparate operating systems and a lack of consistent data models across networks, delivering systems with this level of visibility has been impossible.

Arista’s development of CV UNO is a welcome addition for IT groups challenged by stringent regulatory compliance, cybersecurity and observability throughout the enterprise. These kinds of tools are no longer optional but rather essential imperatives. Automation, once a dirty word in engineering circles, is now critical to keeping networks up and running, and this starts with visibility. The adage goes, “You can’t manage what you can’t see,” so it stands to reason the more you see, the more you can manage.

Zeus Kerravala is a principal analyst at ZK Research, a division of Kerravala Consulting. He wrote this article for SiliconANGLE.

Image: Arista