Accelerate Deployment of Interactive and
Immersive Media Services with Intel’s Converged
Edge Media Platform
Media services have been growing exponentially over the last decade, with video traffic accounting for >80% of consumer internet traffic by 2022.1 This decade will see visual services evolving from passive content to become more intelligent, interactive and immersive. These new services are latency-sensitive and bandwidth-intensive, and they will require video content to be processed closer to the end-user at the edge. No wonder the edge computing market is expected to grow from $10B in 2020 to >$400B by 2030, with the largest part by far made up of video.2 This presents a significant opportunity for telcos who have so far missed out on cloud growth. The unparalleled reach of their network edge assets and visibility into network traffic provides telcos a huge competitive advantage to benefit from edge computing growth. But predicting which video services will take off in the coming years is extremely challenging, and no one can afford to invest in technology dead ends.
Intel’s Converged Edge Media Platform helps overcome that risk by providing a cloud-native platform optimized for dynamically deploying multiple video services in power- and space-constrained edge environments. Telcos no longer need to place bets in the form of dedicated resources for services that may not flourish. Instead, they can build on a general-purpose architecture where services reside together and can leverage cloud-native scalability to dynamically change or resize services to changing needs.
The first demonstration of the Converged Edge Media Platform is ready, and we’re showing it off at MWC Barcelona 2023. The development of solution blueprints for specific workloads and combinations of workloads (such as cloud gaming, CDN and video analytics) is well underway. Telcos and others will use the hardware and software specifications and configurations in those blueprints to build out their own video environments at the network edge, which will inherit the fundamental benefits of Intel’s Converged Edge Media Platform:
- Utility: Power- and TCO-optimized implementation in a space- and power-constrained edge environment.
- Scalability: Cloud-native capabilities to scale up and down per peak performance requirements, multi-services from multiple ISVs and service providers.
- Viability: Ready-made ecosystem and Intel-developed reference architectures addressing the most relevant edge media use cases and unique business models for CapEx investments.
Meeting Multi-Service Challenges at the Network Edge
Placing multiple video workloads together at the network edge is deceptively complicated, and that will become more true as media workloads continue to get more immersive and interactive. We needed to define optimal hardware stacks across all the known and unknown possibilities for media workloads. We needed to ensure that GPU-centric use cases such as cloud gaming and analytics can coexist with CPU-centric workloads such as CDN, while getting the best performance across the board. And we needed the environment to scale while protecting SLAs.
Converged Edge Media Platform lies at the core of ecosystem solutions to address those challenges. As a result, telcos will be able to simultaneously provision multiple video services that have different platform requirements, in a single environment. The platform will seamlessly meet heterogeneous requirements for CPU, GPU, storage and network resources, automatically and intelligently scaling up and down as needed for usage peaks. At the same time, the environment will provide robust support for meeting and validating different SLAs, with capabilities such as mediation for noisy neighbor scenarios. This video-optimized network edge is cloud-native to its core, meaning that it is fully compatible with other modernized infrastructure based on containerized network functions.
The demos at MWC showcase how multi-tenant video services can coexist on shared, cloud-native infrastructure. The scheduler places workloads intelligently, based on gathered information about node features, combined with real-time telemetry. Elastic auto-scaling helps ensure that sufficient resources are available to support those placements, without contention. To protect sensitive data in the multi-tenant environment, hardware and software measures work together to isolate workloads from each other.
Join the Journey to Edge Media Innovation
We are very excited to showcase Intel’s Converged Edge Media Platform with our partners Broadpeak, China Mobile, Cloudsky, ThunderSoft and ZTE at MWC. Our collaborations with partners are demonstrating early value with a new generation of multi-service architectures that intelligently respond to changing combinations of requirements, optimized for the network edge. Engage with us by visiting the partner test case scenarios in the Intel booth at MWC Hall 3 Stand 3E31.
- Demo 1: Broadpeak and Cloudsky are using Intel’s Converged Edge Media Platform to integrate CDN and cloud gaming.
- Demo 2: China Mobile, in collaboration with ZTE Corporation and ThunderSoft, is using Intel’s Converged Edge Media Platform to enable CDN and V2X.
1 Cisco. “VNI Complete Forecast Highlights.” https://www.cisco.com/c/dam/m/en_us/solutions/service-provider/vni-forecast-highlights/pdf/Global_Device_Growth_Traffic_Profiles.pdf.
2 STL Partners, December 2022. “Edge computing market sizing forecast: Second release.” https://stlpartners.com/research/edge-computing-market-sizing-forecast-2nd-release/.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.