Pluribus NOS targets Kubernetes and cloud fabrics

To better manage the distributed and containerized resources of enterprise clouds, Pluribus Networks has significantly improved its fabric switching software. The updates add three fabric monitoring capabilities – FlowTracker, KubeTracker and Virtualized Packet Broker Service – to Netvisor One, the company’s Linux-based virtualized NOS that provides Layer 2 and Layer 3 networks and distributed fabric intelligence . Netvisor One OS virtualizes switching hardware and implements Pluribus’ Adaptive Cloud Fabric software-defined SDN networking package. “Adaptive Cloud Fabric is controllerless and can be deployed in any data center or targeted to specific racks, pods, server farms or hyperconverged infrastructures,” Pluribus Networks explained.

FlowTracker, the first new feature in Netvisor One, adds wire-speed network flow telemetry for every type of flow that can traverse Adaptive Cloud Fabric, including TCP, UDP, ICMP flows, as well as infrastructure services such as DNS, DHCP and NTP. Until now, Netvisor ONE and Adaptive Cloud Fabric could only capture TCP flow telemetry. The software collects network telemetry data from Ethernet switches built on Broadcom’s Trident3 and Trident4 programmable Ethernet switch chips. Major network vendors including Cisco, Juniper, Arista, Dell, HPE and others support these chips. “This built-in telemetry support eliminates the need for external test access points (TAPs) and an aggregation TAP overlay infrastructure,” wrote Alessandro Barbieri, vice president of product management at Pluribus, in a blog post. TAPs are hardware devices placed on network segments to monitor network traffic. The cost of TAPS can be considerable and requires cost-benefit trade-offs that often result in the installation of TAPS at only certain points in the network. “FlowTracker avoids this expense, capturing flow metadata at each switch and exporting it to tools like Pluribus’ UNUM Insight Analysis to spot and fix problems faster,” Barbieri said.

An API for Kubernetes

KubeTracker, another feature in the Netvisor One update, monitors and manages the visibility of traffic flows between containers. It correlates containers with applications, determines which hosts they reside on, and how they are connected to the fabric. “The goal is to quickly identify and resolve application availability issues,” Barbieri said. Until now, NetOps teams had no visibility into these flows, making it difficult to verify application availability and network performance for containerized workloads. “The KubeTracker service is supported on any switch in a Pluribus fabric (enabled on any node in the fabric) to subscribe to events published by the Kubernetes API server and ensure that all of the fabric knows how Kubernetes microservices are distributed across the network and how their deployments dynamically change over time,” said the VP of Product Management at Pluribus.

As for Virtualized Packet Broker, the third new component of Netvisor One, it aggregates, filters and replicates any flow coming from anywhere in the fabric and directs this data to one or more monitoring and security tools. “An interesting point to note is that Pluribus can provide Virtualized Packet Broker as a service of the Adaptive Cloud Fabric,” Ron Westfall, Principal Analyst and Research Director at Futurum, wrote in a blog post about Pluribus announcement. “This is key because packet brokers are typically deployed as part of a parallel surveillance network made up of a lot of hardware, including TAPS, which literally taps into the network to create streams of mirrored data, and TAP aggregators that forward aggregated traffic streams from these TAPS to dedicated packet broker devices. This infrastructure can be extremely expensive to purchase, deploy and maintain. Pluribus provides this virtualized network packet broker as a virtual service running on the same fabric that carries Layer 2 and Layer 3 production network traffic. From my point of view, this solution clearly differentiates itself from other solutions, including existing TAPS implementations,” added Ron Westfall.

In addition to these service updates, Pluribus said Netvisor can now run on Dell’s PowerSwitch 100/400Gbe Z9432F-ON, intended for high-end data center network aggregation tasks. Dell and Pluribus have worked together for several years to integrate data center networking hardware and software that competes with offerings from HPE, Cisco and others. Pluribus Netvisor ONE OS for Dell PowerSwitch is available directly from Dell.