The Linux Foundation Projects
Skip to main content

Join us at ONE Summit 2024 in San Jose and explore the latest in networking and edge innovation | REGISTER

By: Beth Cohen, Verizon; Nick Chase, Mirantis

So what is CNTT and why is it so important for the telecom industry?

When I was first invited to join in the effort to define a reference model for architectures to support NFV workloads, I questioned the sanity of undertaking what seemed to be a herculean task.  Did we really need yet another open source project that might or might not go anywhere? Who was going to support this effort?  Did a technology gap really exist?  Who cares about reference models, architectures or standardized infrastructure implementations anyway? 

Well one year later I can confidently say, yes lots of folks in the telecom industry (vendors and operators alike) are passionate about infrastructure. The Common iNfrastructure Telecom Taskforce (CNTT) is a cooperative group supported by LFN, OPNFV, and GSMA, that has tasked itself with defining a reference infrastructure for telcos and other communication service providers. CNTT is real and with the May 2020 Baldy release, the group has done some amazing things in defining reference infrastructure models that we hope will become widely adopted across the industry.  

CNTT was organized as a taskforce under the LFN umbrella to determine a set of common requirements that would enable telecoms and other large infrastructure providers to create a single infrastructure to host multiple Virtual Network Function (VNF) or Cloud Native Network Function (CNF) workloads, avoiding the common problem of needing to deploy multiple expensive infrastructures for each new workload.

CNTT currently has three main workstreams: one focused on virtualized workloads, based on OpenStack, the second focused on containerized workloads using the Kubernetes toolset, and finally, the latest, an Edge workstream. The overall project started with a Reference Model to which all future reference architectures must conform, this limited set of Reference Architectures then define the details, which are used to create the Reference Implementations that are used for testing workloads and Reference Conformance tests for both the infrastructures and workloads.

The most exciting part for me is that Baldy includes the first release of CNTT’s container-based reference architecture (RA-2).  Yes, it also has final conformance requirements for its VM-based architecture (RA-1) and the start of container-based conformance tests, all of which will enable OPNFV’s OVP project to begin testing against the CNTT requirements by the end of the summer. Some of the highlights of the Baldy release include:

  • The first release of the container-based Reference Architecture is now available, defining Kubernetes-based architectures
  • The Reference Conformance suite for VM-based infrastructures has been finalized, enabling OPNFV’s OVP project to start testing against CNTT requirements as early as this fall
  • The framework for the container-based Reference Conformance suite is now available, enabling developers to begin adding appropriate tests and code

The CNTT community is also ramping up infrastructure and workload-based field trials, which are designed to uncover any gaps in the Reference Model, Reference Architecture or Reference Conformance suites. The first testing will take place at the ETSI PlugTest in June.

So if you think that you or your company can benefit or lend expertise to this effort, I encourage you to learn more with the CNTT webpagewhitepaper, and wiki to start contributing to the project. All are welcome!

Author