Understanding the High Availability Feature in NetVisor OS

High availability (HA) refers to the characteristic of a system (an individual device or a group of devices) to ensure that service availability (i.e., uptime) is higher than normal and hence downtime is minimized. In case of networking, this is achieved by increasing the resilience (i.e., fault tolerance) of the network so as to be able to provide and maintain an acceptable level of service during faults and challenges to normal operation.

High availability is generally achieved through multiple strategies that include:

• Redundancy of device components (such as power supplies and fans) to maximize the reliability of individual network nodes.
• Device and link redundancy to maximize end-to-end service continuity by eliminating single points of failure and guaranteeing fast re-convergence.
• Simplified operation as well as network automation and orchestration to minimize potential service-disrupting human errors.

Arista's Unified Cloud Fabric control plane provides intuitive and less error-prone mechanisms to automatically configure and manage a number of physical or logical components as a virtual functional super-entity referred to as a fabric). For details, see the Configuring and Administering the Unified Cloud Fabric chapter.

Moreover, as part of the redundant fabric architecture, the NetVisor OS software supports additional advanced capabilities to increase network automation, resiliency and scalability, such as automatic link aggregation and switch clustering with vLAGs.