Any Tap, Any Tool, Anywhere
As applications and workloads become more distributed across data centers, edge locations and public clouds, a well-constructed, highly available network to connect those locations are becoming the lifeline of digital business. To ensure that this network lifeline performs consistently and securely, IT teams need a cost-effective way to plan, monitor, and troubleshoot the physical and virtual network infrastructure across geographically distributed sites. A new architectural approach to network packet broker services is the answer.
Network Packet Broker Challenges
To ensure smooth functioning of their IT infrastructure, networking, security, and operations teams typically need a Network Packet Broker whose role is to intelligently “broker” (replicate, filter and route) traffic from the production network to various monitoring and analysis tools.
There are a number of Network Packet Broker solutions in the market, each with its own unique approach, but most of them have drawbacks for the emerging application environment of compute resources distributed across primary and secondary data centers, public cloud, and edge locations. At Pluribus, we believe that IT teams need a new architectural approach to packet broker services in order to effectively monitor this distributed infrastructure, one which is dynamic, distributed, highly-available, and cost-efficient.
Architectural Approach to Distributed Packet Broker Services
What is the Pluribus Network Packet Broker (NPB)? It is the industry’s first dynamic network packet broker fabric solution that is simple to deploy, scales out to seamlessly monitor geographically distributed data center networks, and can rapidly adapt to changing network conditions, all implemented on cost-effective open networking Ethernet switches. Because the Pluribus Network Packet Broker is built on the architectural foundation of the Adaptive Cloud Fabric, it inherits many advantages that are valuable for network monitoring and packet broker applications.
The solution can be deployed out-of-band like any legacy monitoring fabric, operating as a dedicated monitoring fabric separate from the production network. The out-of-band monitoring solution (as shown in Fig.1.) can be deployed on white box switches and scale out to thousands of ports. Any physical topology is supported: leaf-spine, ring, hub-and-spoke, tree, full mesh, partial mesh, and any combination of these. The solution can dynamically load balance traffic across multiple links for efficient resource utilization and multiple resilient connections inbound from TAP/SPAN sources and outbound to multiple tools.
A Pluribus Network Packet Broker fabric can seamlessly span multiple sites, allowing tools to be placed either in a centralized pool or distributed without restriction across multiple locations (as shown in Fig.2), and deliver completely automated and rapid sub-second re-convergence. Even across an existing IP Core / WAN distributed network the Network Packet Broker fabric will seamlessly automate the deployment of packet broker policies across multiple locations with a single operation. And of course, path computation, replication, and load balancing are all operations automated in hardware while interoperating with the IP network.
Why Pluribus Network Packet Broker?
The figure below highlights three categories of differentiation that make the Pluribus Network Packet Broker a great choice for emerging distributed packet broker infrastructures.