Adaptive Cloud Fabric: Simplifying the Software Defined Data Center

Today we announced the next-generation of Software-Defined Networking with the introduction of our Adaptive Cloud Fabric™, built on the battle-proven Netvisor® Operating System (OS). The previous generation of SDN architectures have been built on the Centralized Control Plane model, which inhibits architectures with slow reconvergence, single-points-of failures and is really only suitable for small deployments, which didn’t require interoperability. Some of the earlier SDN products were based on Open Switches, while older vendors took a proprietary ASIC-based approach with proprietary protocols that limited the deployments.  But the one thing they all had in common, is that they required a controller, which limited SDN success.

Distributed Software Defined Fabric Architecture

The Pluribus Netvisor® OS runs on OPEN switches and implements a fully distributed, peer-to-peer control plane that interoperates with other switches using standard L2/L3 protocols.

Figure 1: Pluribus offers highly available, interoperable IP underlay with a choice of underlay architecture.

A collection of switches running the Netvisor software, typically in a leaf/spine pod, form the Fabric that appears as a logical switch to 3rd party controllers like Openstack/ODL etc. via Fabric-wide REST API or per-tenant Neutron/OVSDB plugins.

Highlights of the Adaptive Cloud Fabric are:

  • Runs on OPEN switches with off-the-shelf chips like Broadcom Trident/Trident2/Tomahawk, Intel, and other newer chips that are coming online
  • Fully distributed, peer-to-peer control plane with high availability and sub-second convergence for scale and resiliency
  • Abstraction of the underlying switch/port by creating a fabric vPORT that is the virtual endpoint for Server Virtual Machines (VM), Containers, SR-IOV enabled VMs, Bare Metal, IOT devices, etc.
  • Integrated Application Level Analytics without TAPs or mirrors that show every TCP connection including time started/ended, IP addresses and TCP ports of end-points, average latency, bytes in either direction, etc. The data is available in both real time and historical and is stored in memory
  • Fully multi-tenant in the form of VNETs which allow each tenant to work with its own Openstack or VMware environment, create server-to-switch tunnels, configure its fabric-wide ports, VLANs, etc.
  • Fully automated, programmable and interoperable in multi-vendor environments

Figure 2 – The Pluribus Adaptive Cloud Fabric abstracting physical ports to virtual ports (vPorts), abstracting the network hardware directly.

 

Evolution of the Virtualization Centric Fabric  to Adaptive Cloud Fabric

 

Netvisor has been shipping in beta since 2013, and we announced the Virtualization-Centric Fabric at the end of 2015. Netvisor is deployment proven with more than 100 deployments in 20 countries in very demanding leaf/spine pod environments. As we have evolved our architecture, we enabled the ability to span across geographically distributed datacenters that can be connected into a seamless fabric using VXLAN tunnels over any WAN or dark fiber to optimize performance, latency and support millions of concurrent connections allowing organizations to build software defined private or public clouds that can support mission critical virtualized applications, hybrid IT, Internet of Things (IOT), and next generation NFVi applications.

Given the flexible nature of the Netvisor software and its deployment in Software- Defined Infrastructure environments, Adaptive Cloud Fabric is a more apt name going forward.

Figure 3 – Pluribus Adaptive Cloud Fabric provides the ability to span across geographically distributed datacenters that can be connected into a seamless fabric using VXLAN tunnels over any WAN or dark fiber.

Summary

The Pluribus Networks Adaptive Cloud Fabric builds on the deployment-proven Netvisor OS to enable organizations to leverage the power of SDN into Software-Defined Private and Public clouds while providing a very high degree of resiliency, sub-second convergence, multi-tenancy with full application visibility.


About the Author

Sunay Tripathi

Sunay Tripathi

Sunay is the CTO and a Co-Founder of Pluribus Networks. Prior to Pluribus, Sunay was a Senior Distinguished Engineer for Sun Microsystems, and was the Chief Architect for Kernel/Network Virtualization in Core Solaris OS. Sunay has an extensive 20+ year software background, and was one of the top code contributors to Solaris. Sunay holds over 50 patents encompassing network and server virtualization.