3 Reasons To Use Composable Network Protocols From Metaswitch

Composable Networking is an important step towards complete network disaggregation for any network operator. While the disaggregation of software from hardware is now commonly understood and enshrined in the principles of today's white box switching approaches, it is only an initial stage of communications infrastructure evolution. To achieve greater cost efficiency and agility, network architects will want to embrace an approach whereby the individual white box software components are also completely disaggregated. This is what we call Composable Networking.


Metaswitch’s suite of Composable Network Protocols (CNPs) offers fundamental building blocks for composable networks. The CNPs enable a new generation of completely disaggregated hardware, network operating systems and control plane routing applications. They also give network engineers unprecedented control of their routing infrastructure and the flexibility to choose the best routing and control plane components for their specific needs.

To illustrate, here are three ways that Metaswitch CNPs make network operations more efficient and reliable:

1. Hitless, In-Service, Control-Plane Upgrades

One of the biggest benefits of leveraging Metaswitch’s CNPs is that control plane functions can be upgraded without affecting services.

We’ve demonstrated hitless upgrades of the Open Shortest Path First (OSPF) IP routing protocol, as an example. OSPF is a link-state Interior Gateway Protocol (IGP) for the Internet that is used to distribute IP routing information throughout a single Autonomous System (AS) in an IP network. In the short video we demonstrate how network operators can quickly upgrade the OSPF stack to a version that supports new features with zero packet loss and without disturbing existing configuration parameters.

We performed the upgrade by employing a standard Ansible Playbook, but Command Line Interface (CLI) can also be used. The total time it takes to perform the upgrade is just a matter of a few seconds.

2. Simplified, Reliable BGP Monitoring

Network architects must continually scrutinize Border Gateway Protocol (BGP) routing infrastructures to ensure optimal operations and performance. But getting the information they need to make intelligent diagnoses and quickly perform the right corrective and preventative actions has been almost impossible. Traditional solutions, such as BGP route servers, have only limited information. Furthermore, this data is typically gathered through polling using Simple Network Management Protocol (SNMP) or Command Line Interface (CLI) scraping, which degrade the router’s performance.

Metaswitch’s suite of CNPs overcomes these limitations by supporting the BGP Monitoring Protocol (BMP). BMP dynamically streams every significant BGP event without any performance degradation. In this video, we Network operators can get a complete real-time view of BGP state of every router. This data can then be used to create custom monitoring applications.

3. Open API for Streamed Access to Routing Table

Via an open Application Programming Interface (API), network operators can receive streamed programmatic access to the routing table. This enables granular policy-based control of the underlying forwarding infrastructure by centralized orchestration and automation engines that augment classic route discovery mechanisms.

In this example, we simulate a critical application receiving data flowing through two routers. When one of the interfaces to a router goes down, the route controller application is notified so that it can push a new specific route, for the critical application, over an alternative interface.

Metaswitch’s API is highly abstracted and based on GRPC, which enables customized routing logic and also becomes the foundation for easily adding third-party protocols.

Click here to learn more about our portfolio of Composable Network Protocols.