The Adaptive Cloud Fabric™ (ACF) delivers new levels of network underlay and overlay automation, traffic visibility, and network segmentation in an extremely cost-effective solution that works right out of the box. Built on the deployment-proven Linux-based Netvisor® ONE network operating system (OS) and virtualization software, the Adaptive Cloud Fabric, an advanced data center fabric solution, is a simple, fully automated, virtualized and secure approach to building a distributed network architecture that brings the benefits of cloud-scale, elasticity and adaptability to the modern data center and campus distribution networks. Unlike more expensive and complex controller-based software-defined networking (SDN) solutions, the Adaptive Cloud Fabric is a controllerless SDN solution that automates and integrates the physical data center network underlay and virtualized network overlay into a unified networking fabric. ACF can automate and virtualize the network for a single data center site or it can be easily stretched across multiple data center sites, including highly distributed edge compute environments, to deliver a single unified and programmable network fabric.
- A single command programs the entire fabric; for example add a fabric-wide VLAN or subnet or shut down a port on all nodes with a single command.
- Every command is atomic with unlimited rollback to quickly revert to a good state.
- VXLAN tunnels are automatically established across all nodes in the fabric for the overlay along with automated VLAN<>VNI<>VTEP mapping to reduce complexity.
- Deep slicing (network segmentation) provides isolation for specific traffic types like IoT as well as supporting multi-tenant services.
- Rich and highly scalable layer 2 and layer 3 (IPv4/IPV6) VPN services can be deployed across the entire fabric on a per slice basis.
- Integrated telemetry captures all key parameters of every TCP flow between all endpoints connected to the fabric for unparalleled visibility and analytics.
- Supports any topology including leaf/spine, ring and other architectures.
- Can be flexibly deployed as a standard OS on a single switch, as an automated underlay fabric supporting compute-based overlay virtualization solutions or as an integrated, turn-key automated underlay/overlay fabric solution.
To meet the most stringent high availability requirements, the Adaptive Cloud Fabric architecture has no single-point-of failure and delivers a high degree of resiliency with fabric-wide sub-second failover. Fabric automation provides a single-point-of-management and control, distributes intelligence, integrates a broad range of advanced L2 and L3 network services, and provides pervasive visibility for all traffic traversing the fabric. The Adaptive Cloud Fabric interoperates with existing network infrastructure, is highly scalable and is optimized to deliver continuous availability for mission-critical enterprise and service provider environments. ACF can scale-out to support many thousands of ports, with multi-terabit capacity, performance and latency predictability, and support millions of concurrent connections.
Adaptive Cloud Fabric controllerless architecture is achieved via a distributed software model where the SDN control intelligence and fabric-wide state database is efficiently run as an application in the user space of each white box networking switch that is being deployed for physical underlay connectivity in the data center or campus distribution network. The highly distributed software is designed to leverage the server-like characteristics of modern white box switches including powerful multi-core CPUs and packet processing ASICs along with RAM and SSD memory footprint. Once ACF is deployed, the group of switches, even across multiple data centers, federates together into a peer-to-peer fabric to automate both the underlay and the VXLAN overlay. The fabric is completely unified and treated as one logical, programmable and software-defined switch-router that is easy to manage using open SDN automation via CLI, RESTful API or our graphical UNUM Management Platform. This distributed SDN approach maintains the use of standardized protocols such as BGP, OSPF and EVPN which enables seamless interoperability with existing infrastructure.
Pluribus Networks – Networking. Simplified.