close search bar

Sorry, not available in this language yet

close language selection
 

CORD Project: Driving network solutions with open source

The CORD® platform uses leading-edge SDN, NFV, and cloud technologies to build nimble inline data centers at the edge of operator networks. Learn more.

CORD Project: Driving network solutions with open source

CORD® (Central Office Re-architected as a Datacenter) is a platform leveraging leading edge SDN, NFV and Cloud technologies to build nimble in-line datacenters at the edge of operator networks. CORD integrates a curated collection of dozens of leading open source projects, thus making a fully-integrated platform for building innovative solutions available for network operators. Designed to leverage the best of modern DevOps application development methodologies, CORD delivers an open, programmable, agile platform for service creation.

Much of the success of CORD is driven by the various pre-integrated releases based on CORD to deliver solutions for various operator use cases. Solutions available today include Residential (R-CORD), 5G Mobile (M-CORD) and Enterprise VPN (E-CORD). So significant is the early success of the project, IHS reports 70% of operators plan to deploy CORD in their central offices by 2018.

How did the CORD project get started?

The CORD Project was launched by Open Networking Lab (ON.Lab), a non-profit organization fostering open source communities for developing tools and platforms to realize the full potential of software-defined networking. Initially developed as a use case for the open source SDN operating system ONOS®, CORD came about from ON.Lab conversations with AT&T that uncovered where the service providers’ main costs were. ON.Lab looked at the edge and how central offices were being built. We saw a number of limiting factors, such as lack of agility with regard to building new services, and operators unable to move on to compete with over-the-top players. ON.Lab came to understand that in order for service providers to compete, the central office needed to evolve.

AT&T was the first partner to join the CORD initiative within the context of ONOS, building out CORD as a use case. Additional service providers started to join the effort, focusing on their priorities in terms of what would help the industry transition to virtual networks. This turned into a cross-industry effort among service providers, vendors and developers supporting ON.Lab’s mission around transforming the central office.

CORD has experienced major momentum and adoption among service providers and vendors the past two years; so much so that in July 2016, CORD became a separate open source project with its own governance, partners, collaborators and contributors actively collaborating and accelerating CORD development targeting use in production networks.

ON.Lab is merging with the Open Networking Foundation to drive network solutions, like CORD, with open source software and software-defined standards. As a Linux Foundation project, ON.Lab works closely with the organization to ensure open, neutral and merit-based governance of the project.

Was there any true “champion” for the project?

As an independently funded software project hosted by The Linux Foundation, CORD Project is neutral with third-party governance and technical contributions based on meritocracy. CORD is backed by providers like AT&T, China Unicom, Google, NTT, SK Telecom and Verizon, but the project is not dominated by any one group or company.

What makes it different or unique?

CORD allows service providers to fully take advantage of SDN, NFV and cloud with a single integrated solution used to quickly and safely configure, deploy and manage cloud-scale service packages with attractive economics. CORD offers a unified interface, Web-like simplicity, automated provisioning, open APIs, built-in security and the ability to download and auto-build CORD on a single node in about an hour. CORD streamlines operations by breaking down silos between SDN, NFV and elastic cloud technologies. By unifying SDN, NFV and the cloud in a single solution, CORD goes beyond what each technology contributes in isolation.

How did you build your community?

The CORD community was built through building use cases. We started with residential CORD, followed by enterprise and mobile versions of CORD. Each use case attracted both service providers and vendors interested in building and improving the use cases. In addition to developers from partners, dozens of collaborating companies joined to help move the solution toward field trials.

The CORD community also has the benefit of being able to leverage the lessons learned from the ONOS community and that has allowed the CORD community to scale quickly. For instance, ONOS rolled out an initiative called brigades last year that rallied community members around accelerating features on the roadmap. CORD will be adopting that model itself this year to assemble groups of engineers from across the community to focus on building the features.

The CORD community is also leveraging the Ambassador program, which was developed first for ONOS last year. This lets us work with community evangelists around the world who can speak about CORD in their own communities. This program is excellent, because many of these places are areas where we don’t have staff who can do that outreach.

How did you attract such a large project team so quickly?

CORD Project is rapidly growing and diversifying its global community of cross-industry contributors by fostering innovation through open source collaboration. CORD has gained more than 50 partners and collaborators, leveraging the power of open source to rapidly bring use cases and best-in-class solutions for open networking into field trials and deployments. With Web-like simplicity and automated internet-like provisioning, CORD allows subscribers to quickly and safely configure and manage service packages. I think the ease of use really helped to attract people to the project.

What is your philosophy on building community?

Network infrastructure is primed to be in its next generation, most of which runs on commodity type devices and OSS. ON.Lab’s mission has been to use open source collaboration to accelerate adoption of SDN and NFV and bring a better ecosystem for networking to fruition.

Now that the SDN movement, first initiated by the ONF, has successfully set in motion the disaggregation of networking devices and control software and fostered the emergence of a broad range open source platforms, the industry needs a unifying effort to build solutions out of the numerous disaggregated components. A trend has emerged where vendors leverage open source to build closed proprietary solutions, providing only marginal benefit to the broader ecosystem.

CORD is building its community and solutions through the ONF’s Open Innovation Pipeline, which intends to counteract this trend by offering greater returns to members who participate in the ONF’s collaborative process. Through making active contributions to the Open Innovation Pipeline, vendors benefit from inclusion in CORD and ONOS solutions, thereby gaining access to operator deployments. Now, any member with a valuable contribution can insert their unique innovation anywhere along this open pipeline, and the momentum of the pipeline will pull the innovation into operator PoCs, trials and beyond. Operators, vendors and integrators all have a role to play, and the pipeline helps integrate these contributions into consumable solutions for operators.

What other technologies is your project is aligned with?

CORD combines SDN, NFV and cloud technologies with common infrastructure to provide service providers with the open source building blocks to create innovative new services spanning the telco central office, access, home, and enterprise.

What other open source projects are you leveraging to create the project, if any?

At its core, CORD is a curated integration of numerous open source projects, thus creating a platform for innovation. CORD incorporates a broad range of leading edge upstream open source projects to create a next-generation cloud-centric datacenter. The complete range of IT functionality is provided, including compute, storage and networking. This is done by leveraging and integrating the following open source projects:

IaaS

  • OpenStack
  • Ubuntu
  • MaaS

DevOps

  • Docker
  • Ansible
  • Django
  • Angular
  • Juju
  • Vagrant

Networking

  • ONOS
  • OF-DPA
  • ONL
  • xRAN

Service Creation

  • XOS

Challenges

You have an ecosystem where the ecosystem you’re producing is going to impact the people who provide the infrastructure you “used to” use. The equipment is getting disaggregated, the functions on that equipment are getting replaced, and so on. There’s a challenge to help the vendors find new business models but to still be successful in an ecosystem like this.

Opportunities

CORD is available in distributions for three main domains of use:

  • Residential (R-CORD): R-CORD is for residential subscribers connected over wireline technologies, such as GPON, G.Fast, 10GPON, and DOCSIS. R-CORD includes access devices built as white boxes with merchant silicon and a set of services such as CDN, parental control and other services designed for residential customers.
  • Mobile CORD (M-CORD): Now ready for lab explorations, M-CORD is the open reference solution for service-driven 5G architecture. M-CORD features end-to-end open source slicing from programmable Radio Access Network (RAN) to disaggregated and virtualized Evolved Packet Core (EPC).
  • Enterprise CORD (E-CORD): E-CORD offers enterprise connectivity services over metro and wide area networks, using only open source software and commodity hardware. E-CORD builds on that same CORD infrastructure to support enterprise customers. Service providers can continue to offer enterprise connectivity services (L2 and L3VPN), but go far beyond simple connectivity services as E-CORD allows them to include virtual network functions and service composition to support disruptive cloud-based enterprise services.

What are your goals for the upcoming year?

Today’s mobile infrastructure can’t support IoT devices and the constant connection times and huge amounts of video data, or the locality of services you’d like to provide around events. Mobile is going to explode over the next two years with the infrastructure ready to switch to 5G.

The world of telecom has been about standards, and 3-5 years may have been their cycle. With software being the infrastructure and fast POC iterations, we anticipate you’re going to see implementations coming into the field quickly, so there’s a lot of discussion around how software and standards communities can make a positive impact.

2017 is the year of taking CORD out of lab and into field trials. We’ll drive more POCs, and harden infrastructure.

What are your long-term goals for the project?

CORD Project’s mission is to redefine the economics and agility of access networks across the globe. By developing a flexible, agile and open source framework, CORD Project aims to help service providers re-architect their traditional central offices into next-generation data centers.

Learn about other Open Source Rookies!

 
Larry Peterson

Posted by

Larry Peterson

Larry Peterson

Larry Peterson is the CTO at Open Networking Lab/Open Networking Foundation, and Chief Architect of the CORD project. He is also the Robert E. Kahn Professor of Computer Science, Emeritus at Princeton University. While at Princeton, he served as Chair from 2003-2009, led work on the widely used PlantLab and MeasurementLab platforms, and spun out a CDN technology startup that was subsequently acquired by Akamai. Peterson is a member of the National Academy of Engineering, a Fellow of the ACM and the IEEE, the 2010 recipient of the IEEE Kobayashi Computer and Communication Award, and the 2013 recipient of the ACM SIGCOMM Award. He received his Ph.D. degree from Purdue University in 1985. Learn more about the project at http://opencord.org/blog/.


More from Open source and software supply chain risks