Intelligent Peering

Intelligent Peering

Evolve your peering to meet the needs of 5G and beyond

How can Intelligent Peering help improve your network?

Your customers expect connections to work immediately. They don’t want slowdowns, buffering, and pixilation. You can exceed your customers’ expectations with a peering solution that’s designed to deliver real-time flow analytics and dynamic routing controls that use your network more efficiently and minimizes backhauled traffic.

Peering Fabric Design

Evolve your architecture

Globally by 2022, 82% of all IP traffic (both business and consumer) will be IP video traffic and it is expected that 33% of service provider network capacity will be in metro areas1. This traffic strains already busy backhaul routes to aggregated peering points. Cisco Intelligent Peering focuses on localized peering to improve transit operations.

  • Minimizes transit distance and eliminates needless network hops.
  • Maintains throughput and capacity with 24x or 36x –100G per RU high density routers.
  • Embraces model-driven telemetry to collect high-fidelity data to predict and resolve congestion before it happens.
  • Builds trusted network fabrics by protecting against distributed denial of service (DDoS) attacks. It also detects hijacking of border gateway protocol (BGP) routes.
  • Helps enable localized peering, which improves resiliency with redundant peering connections in one location. It eliminates the need to add backbone capacity for failover to other regions.

The benefits of Intelligent Peering

Improves the customer experience

Improves the customer experience

Bring content closer to your clients for less congestion and lower latency.

 Builds resiliency

Builds resiliency

Reduce the impacts of maintenance down time or outages with automatic traffic re-routing.

Lowers total cost to operate

Lowers total cost to operate

Lower CapEx through network optimization and reduced backhaul routes. Reduce OpEx through operations automation, analytics driven growth modeling, and a flexible consumption model for licensing.

Increases network agility

Increases network agility

Dynamically re-route traffic and reduce time to scale for localized demand growth.

Cisco Intelligent Peering Solution elements

Cisco Intelligent Peering Solution elements

  • A simplified control-plane that is built upon IPv4/IPv6 with Segment Routing.
  • Peering fabric spine (PFS) routers use ECMP to balance traffic between peering fabric leaf (PFL) routers and are responsible for forwarding within the fabric and the rest of the provider network. Each PFS acts as a label edge router (LER) that is incorporated into the control plane of the core network.
  • In a collapsed design, each peering node is connected to External BGP (EBGP) peers and upstream to the core.
  • In a distributed design, network functions are separated. Peering termination happens on PFL nodes. PFS aggregation nodes are responsible for core connectivity and perform more advanced LER functions.

Technical documentation

Peering Fabric Design

Peering Fabric Design

Peering fabric that uses high-density and low-footprint hardware requires much less space than older generation modular systems. It removes the need to use node pairing for these functions.

Peering Telemetry

Peering Telemetry

Explore IP network telemetry data types, their usage applications, and their configuration in peering environment software to detect network anomalies and provide meaningful insight into network behavior.

Internet Edge Peering

Internet Edge Peering

Peering is the glue that holds the Internet together. It facilitates the flow of data across networks. Find out more about the technical aspects of peering including peering history and current peering architectures in network, security, and telemetry.

Express Peering Fabrics

Express Peering Fabrics

See how regional media or cable hub routes can connect directly to local peering connections. By bypassing regional core aggregation sites, the routes can operate with a simplified control-plane with less power and memory resources, which saves the operator money.