Transforming Network Infrastructure for Cloud-Optimized 5G Services

SHARE:
Copied!

 

Co-authored by:

Rajesh Gadiyar, CTO, Network Platforms Group, Intel Corp.

Kevin Shatzkamer, Vice President, Service Provider Strategy and Solutions, Dell EMC

Imagine the future – a scalable, composable and automated network infrastructure that meets the high-performance needs of tomorrow’s 5G services and makes it easy to build, deploy, manage, and provide assurance for new end-to-end applications. The new era of 5G networks will see technology and operational transformation, but also business innovations that result in intelligent applications consuming and generating data like never before. These intelligent applications will introduce a new set of requirements – latency, bandwidth, capacity, coverage – that drives a transformation of the entire end-to-end architecture, including the compute, network, and storage infrastructure. What used to be possible only in science fiction movies – flying drones, driverless cars and planes, machine-to-machine interactions, seamless communication around the globe – is fast becoming a reality.

Since 2012, the industry has experienced acceleration of changes leveraging the capabilities of compute, network, and storage virtualization to drive new capital and operational models, deliver new services, and improve overall service delivery economics. The new operational imperatives for Communications Service Providers (CoSPs), can largely be captured into three core technology shifts:

  • Leverage disaggregation of hardware and software stacks to shift workloads towards general purpose compute, such as Intel architecture.
  • Decouple core infrastructure and network services from applications and protocols, and expose those services as a platform to applications.
  • Develop a set of information models, data models and APIs to transform operations from bespoke processes and associated infrastructure scripts, to more unified automation frameworks that allow service providers to develop DevOps-style operational processes.

Every decade, the mobile industry goes through a major upgrade cycle of their network architecture – from the Radio Access Network (RAN) to the Packet Core – to meet the insatiable demand of smart mobile devices and the new generation of applications and services.

The impending 5G transition, with significant advances in bandwidth, latency and quality of service, will unleash a new wave of services including enhanced mobile broadband, connected cars, drones, smart retail, industrial robots and much more.

Why a Software-Defined Infrastructure Is Critical to the 5G Transition

The advent of smart everything and the growing amount of data traffic is already putting immense pressure on the network infrastructure. The CoSPs – aka telco operators – have a business challenge: How do they deal with this increased traffic efficiently at a lower cost and allow for rapid deployment of end to end services to grow business.

This motivates the CoSPs to aggressively transform their network by virtualizing network functions on cloud infrastructure with Network Function Virtualization (NFV) and Software Defined Networks (SDN). The result is a homogeneous infrastructure that utilizes standard high volume Intel servers that deliver TCO benefits as well as the cloud scale necessary for next generation of applications and services. The NFV deployments are already happening, and over the last three years we have seen strong commitment from CoSPs around the globe.

The smart applications enabled by 5G technology will utilize concepts such as network slicing to deliver high bandwidth and lower latency, with end-to-end Quality of Service (QoS), ranging from devices to network infrastructure and the cloud. The need for service assurance drives a few key requirements:

First, the infrastructure needs to be composable and scalable, so the compute resources such as CPU, memory, storage and networking can be dynamically assigned based on the needs of the application.

Second, these resources need to be configured, monitored and managed somewhat “automatically” based on well-defined policies and a Management and Orchestration (MANO) framework based on standard APIs also known as ‘closed loop automation’.

Third, the infrastructure needs to be secure and resilient. As multi-tenancy and service innovation on an open architecture becomes the order of the day, end-to-end security needs to become pervasive.

Intel and Dell EMC: The Pillars of 5G

5G represents a significant opportunity for the telecommunications industry to bring together these Digital transformation trends into a unified architecture leveraging the capabilities that Software-Defined Networking (SDN), Network Function Virtualization (NFV), and increasing amounts of automation and orchestration. The insatiable demand for ubiquitous, uniform connectivity and capacity, enabling services such as the Internet of Things, Connected Vehicles, mission-critical data services, and broadband everywhere, delivered with cloud economics, creates the perfect opportunity for 5G networks to leverage the new technology discussed above.

Intel and Dell EMC are aligned on the pillars of 5G networks, the need to embrace the three core technology shifts, and have an established relationship to research and develop solutions that help CoSPs rise to the challenge by:

  1. Engaging strategically with CoSPs to define, prioritize, research, develop and bring to market infrastructure innovations that solve key challenges the industry is facing.
  2. Developing a cloud optimized, 5G RAN infrastructure with a virtualized Radio Access Networks (vRAN) or Cloud RAN (CRAN), leveraging the benefits of pooled resources for baseband and packet processing.
  3. Developing a joint infrastructure solution for Multi-Access Edge Computing (MEC) allowing intelligence at the edge with new applications such as Augment Reality (AR), Virtual Reality (VR), Analytics at the Edge, Edge Caching for Content Delivery Networks, IoT Gateways, and other low-latency applications.
  4. Creating a joint lab environment where Intel and Dell EMC engineers can collaborate on researching, prototyping, and developing solutions.
  5. Investing in applied research that helps the industry understand the benefits of infrastructure enablers embedded in general purpose computing platforms. This includes leveraging FPGAs for workload acceleration, NVMe technology, DPDK for efficient Data Plane processing, and Intel® Quick Assist Technology (QAT) to deliver end-to-end security, improving the performance and efficiency of workloads at the network edge.

In summary, the world is changing rapidly. The advent of “smart everything” and the intelligent end-to-end applications driven by 5G will require an infrastructure transformation – to an efficient, flexible and composable infrastructure. The Intel and Dell EMC partnership is delivering the solutions to enable this infrastructure transformation and enable this “smart everything” revolution.

Continue Reading
Would you like to read more like this?

Related Posts

Click to Load More
All comments are moderated. Unrelated comments or requests for service will not be published, nor will any content deemed inappropriate, including but not limited to promotional and offensive comments. Please post your technical questions in the Support Forums or for customer service and technical support contact Dell EMC Support.