Reference Architectures and Experience Kits for Cloud Native

With core network infrastructure on a rapid path to becoming fully virtualized with cloud native practices, it’s critical for systems developers to be able to efficiently design, produce and deploy reliable applications and services from myriad software, networking and hardware components.

I’ve been developing networking products for the telecommunications sector for most of my career, starting in Israel and then immigrating to the United States two decades ago. I’ve always had a systems engineering perspective and a passion for helping service providers better understand how they can more easily consume the latest technologies to build their applications and services.

In my most recent role, I was faced with the challenge of how to help communication service providers (CoSPs) accelerate the design and deployment of applications and services running on virtualized, multi-vendor solutions tailored for their unique operating environments. These service providers want to take advantage of the latest-generation platforms and open source software innovations.

Collaborating with the CNCF Community

My team’s participation in the CNCF Telco Group, CNF Test Bed and various Kubernetes workgroups and Special Interest Groups (SIGs) for resource management, network, node scheduling and instrumentation has helped enormously to inform our thinking.

The solution lay in reference architectures (RAs) and Experience Kits – essentially blueprints and guides for network deployments. These RAs are comprised of open source software components configured in an optimal way, with many knobs (config options) to make the CoSP’s service or application run well. The goal is for the RA to be a forward-looking blueprint for deploying new technologies at scale across environments and use cases.

The RAs are then complemented by Experience Kits with best-practice guidelines for how to architect, automate, optimize and adapt the network platform for cloud native software technologies such as Kubernetes, service mesh, security, packet processing, telemetry/observability and sustainability.

Delivering to Specific Requirements of Network Applications and Services

We set out to specifically address applications and services delivered across the network from on-premises, to the core network, and to the cloud. Depending upon where the application runs, there can be different requirements and different software projects which come into play. For example:

1. Optimizing power management for performance or for energy efficiency.
2. Guaranteed performance and latency requirements at the edge (Data Plane Development Kit or DPDK).
3. Observability for network health visibility and service assurance (CollectD, Telegraf, OpenTelemetry).
4. Consolidated Kubernetes clusters for scheduling and orchestration of workloads at any location in the network.
5. Helm charts and Kubespray for ease of deployment (K8S operators, Helm, Ansible).

The Cloud Native Computing Foundation (CNCF) hosts critical components of cloud native software stacks, including Kubernetes and Prometheus. CNCF serves as the neutral home for collaboration and brings together the industry’s top developers, end users and vendors.

Learn More

The latest from CNCF

Reference architectures and Experience Kits bring together an ecosystem of partners offering a comprehensive and open platform for industry collaboration and innovation.

For examples of reference architectures and Experience Kits, visit the Intel Network Builders site.

To learn more about networking and other cloud native technologies, consider coming to KubeCon+CloudNativeCon North America 2021 on Oct. 11-15.

Dana is a product management director for Network and Cloud Edge at Intel. She has deep technical and business acumen in the wireless and telecom networking arenas and has collaborated with communities on technology initiatives such as SDN/NFV, cloud native,...

Read more from Dana Nehama