Organizations need agile, scalable networks to meet the increasing demands of new and evolving use cases, from streaming media to industrial robots to intelligent traffic management, and to deliver the highest-level quality of service. Pushing compute to the network edge, closer to where data is created and consumed, provides the speed and actionable insights you need for monetization opportunities and operational savings.
The Value of an Agile Network Edge
Service providers face the daunting tasks of delivering competitive, differentiated services, meeting the performance demands of today’s data-intensive and latency-sensitive applications, and positioning themselves to quickly adapt to evolving standards to offer the innovative services of tomorrow.
Existing networks that primarily rely on a consolidated cloud computing and data storage model or on fixed function physical infrastructure are challenged to economically deliver the near real-time responsiveness required by existing use cases. They are further challenged by the scalability needed to support the rapidly growing number of sensors and devices, estimated to reach 50 billion by 20221, and data volumes of the near future, predicted to grow to 175 zettabytes by 2025.2 3
In order to respond to customer expectations while controlling capital investment and operational cost, network infrastructures need to be transformed to be agile, scalable, and intelligent. They need to adjust to meet variable system demands, providing lower latency, improved throughput, and greater reliability, and to enable capitalization of every data point in the network, leading to new revenue stream opportunities and insights.
Essential to this modernized infrastructure is combining the power of edge computing—for data computation, data analysis, and data storage at that point of service delivery or origin of data creation and consumption—with cloud-ready software infrastructure running on shared physical resources for optimal workload placement for highest return.
Benefits of Computing at the Network Edge
Placing compute at the network edge makes it possible to intelligently manage workload placement and provides several key operational benefits:
- Increased speed and lower latency. Moving data processing and analysis to the edge helps speed system response, enabling faster transactions and better experiences that could be vital in near real-time applications.
- Improved network traffic management. Minimizing the amount of data sent over the network to the cloud can reduce the bandwidth and costs of transmitting and storing large volumes of data.
- Greater reliability. The amount of data that networks can transmit at one time is limited. For locations with subpar internet connectivity, being able to store and process data at the edge improves reliability when the cloud connection is disrupted.
- Enhanced security features. With proper implementation, an edge computing solution may help increase data security by limiting the transmission of data over the internet. In healthcare, for example, there may even be local or regional requirements to limit the storage or transmission of personal data.
- Agility and scalability. With the ability to strategically distribute computing across the network, companies can ensure applications settle in the place of highest value and make the best use of their resources. This opens up a range of opportunities, from easily delivering new services and shifting service delivery between localized markets, to effortlessly expanding capacity without the need to open a new data center.
Considerations for the Network Edge
The network edge sits within your network just outside the network core and includes converging locations such as regional data centers, next generation central offices (NGCOs), fixed wireline access points, wireless access base stations, and radio access networks (RANs). It can also include on-premises locations, such as universal customer premises equipment (vCPE) devices where multiple workloads such as software-defined wide area network (SD-WAN) and enterprise applications can be hosted in a single edge-computing device.
Transformation of the network edge is about system optimization and increased control over traffic management, eventually to the point where you have control over how and where to process and transmit data at the data-packet level based on its value.
Service providers have already adopted software-defined networking (SDN) and network functions virtualization (NFV) to bring flexibility into their networks, but this approach can be extended further. Adopting cloud-like management practices and implementing a software-defined edge can bring increased operational agility and resource efficiency.
Combining regional fixed and mobile network functions into a new central office, deploying private wireless networks, or ensuring secure multi-cloud enterprise vCPE solutions will take a careful balance of performance and agility for cloud-native delivery of rigorous edge services.
Service providers and industries are taking advantage of Intel® technology and expertise for network edge deployments. Intel helps bring together the products, ecosystem, and expertise to accelerate commercial solutions at the edge.
For example, Verizon partnered with Intel, Samsung, and Wind River to deploy the world’s first fully virtualized end-to-end 5G data session in a commercial network, proving the readiness of virtualized RAN technology.
Intel also helped Lumen (formerly CenturyLink) develop a network that delivers both high performance and power efficiency. Working closely with Intel and using Intel® Xeon® Scalable processors, Lumen has added flexibility to its service delivery by customizing performance within power footprint limits.
Intel® Solutions for Network at the Edge
Intel offers a full suite of silicon, software, and tools for network edge infrastructure to meet your cost, performance, and time-to-market needs.
- Drive down long-term costs while delivering edge service innovation. Create an agile, performant edge network across RAN, enterprise on-premises edge, central office, and regional data centers. Blend off-the-shelf and network-optimized CPUs with a wide range of in-network ASIC and SmartNIC accelerators, including structured Intel® eASIC™ devices and fully programmable Intel® FPGA SmartNIC platforms. Convert any location into a software-defined micro data center with our power-optimized, extended-temperature, and long-life offerings in hyperconverged enclosures that deliver on the most-demanding visual, AI, and sub–20 millisecond latency services. Explore solutions.
- Affordable, innovative high-performance networking. Novel technologies like Precision Time Protocol in the Intel® Ethernet network adapter lineup bring higher accuracy IEEE 1588 PTP synchronization signaling to Ethernet and onto the edge platform with lower TCO than existing hardware appliance–based approaches. Intel® Tofino™ programmable switches help offload CDN services and support network slicing. Edge-native 100G silicon photonics transceivers in base stations easily scale for demanding visual and AI services.
- Get to market faster with a mature ecosystem of certified vendors and solution. Intel® Network Builders enables over 400 members to deliver optimized workloads, software platforms, and emerging use cases for network-edge demands. These solutions lower the barriers to containerized network function deployment. Intel® Select Solutions for network also offer a faster and more efficient deployment path for verified, reliable infrastructure configurations that support virtual network enhancements.
- Rigorous network security innovation to stay ahead of new threats at the edge. Technology designed for network security continues to advance, including FPGAs with Intel® Platform Firmware Resilience (Intel® PFR) and Intel® Ethernet network adapters with platform resiliency features, which greatly expand BIOS and Firmware image protection. New Intel® Software Guard Extensions (Intel® SGX) capabilities help secure data in use, not just in flight or at rest.
- Open software for access to all network control points. Open Network Edge Services Software (OpenNESS) is an open source MEC toolkit that enables highly optimized and performant edge platforms to onboard and manage applications and network functions with cloud-like agility across any type of network. The Converged Edge Reference Architecture (CERA) can also be used to help speed up time to market resulting in readily deployable, vertical-aligned service platforms. Intel is also a leading contributor to a number of open-standard organizations, fostering industry-wide standardization to propel networking forward.
Intel Powers the Network Edge
The unprecedented demand on existing networks to deliver exceptional customer experiences and innovative services necessitates a shift to a more agile, scalable network infrastructure. Intel offers a fast, effective path to design and deploy solutions to meet business needs for cost, performance, and time to market.