The power of edge computing

The Multi-access Edge Compute (MEC) framework enables mobile operators, application developers, and content providers to deploy predictable cloud-computing capabilities at the network’s edge and in the immediate proximity of mobile networks. As the industry moves into subsequent phases of the 5G rollout, the MEC framework will be a key enabler for operators that face the need to deliver a greater number of new services, with lower latency and strict Quality of Service (QoS), at higher speeds, over more endpoints than ever before.

MEC provides a framework for multiple organizations to interact with an operator’s Radio Access Network (RAN) as well as a community of other organizations, all in a secure multi-tenant environment. The framework allows an operator to authorize and create contained environments for third-parties in a standardized way that facilitates ease of integration, deployment and lifecycle management. In doing so, it enables operators to offer new services, as well as the ability to sell their hosting capabilities and 5G connectivity to Over The Top (OTT) providers and enterprises. Furthermore, the right combination of operator and application providers enables these coalitions to innovative much faster, capturing market opportunities ahead of the competition.

Being at the network’s edge provides data locality for instant action. It is ideal for low latency and mission-critical services. The MEC architecture provides an interoperability framework that enables software applications to use real-time information describing local network conditions.  In turn, the local environment can be fed back into the operator’s network, thus creating synergy between the application, the network and the user. The ability to predict and react to local data and network states can then be used to influence the user’s overall Quality of Experience (QoE), which ultimately leads to the success of the service.

MEC – 5G service revenue opportunity

The majority of the MEC opportunity is driven by the maturity of 5G deployments. There are a number of ways to quantify the MEC forecasts. According to STL Partners[1] the market for MEC is to be between $1.75G and $4.25B in 2025. The key technologies fueling this growth include Virtual Reality (VR) and Augmented Reality (AR) gaming, seamless video calling, Ultra-High Definition (UHD) videos, and the Internet of Things (IoT) that spawn SmartX applications, including cities, manufacturing, agriculture, and logistics. Regardless of MEC’s proportion of the overall 5G services revenue, the two will grow hand in hand. The new capabilities of 5G hold the potential for significant revenue growth over the long haul – far more than 5G connectivity alone can deliver.

[1] https://stlpartners.com/articles/edge-computing/mobile-edge-computing/

Edge Challenges

Although there are useful edge frameworks such as MEC, that doesn’t mean that every challenge is solved.  There are thousands of different possible topologies from fully distributed federations to hierarchical hubs and spokes.  Furthermore, there are numerous applications, each requiring different user expectations, as well as multiple cloud technologies, like VMs, containers and physical devices.

To effectively enable this melding of multi-tenant applications and infrastructure at large scale, one must:

• Automate 10,000s (bare-metal servers to NFs, applications and services) of edge nodes and unify them back to the core cloud
• Eliminate resource silos with the capability to share resources among multiple cloud technologies
• Eliminate operations silos, a unified operations model is required under the cover, not just a single-pane-of-glass administrative GUI, multiple cloud platforms should not be necessary.  For example, VMs and containers
• Effectively monitor the entire solution as a whole and correlate that data to actionable policies
• Allow multiple organizations and DevOps teams to collaborate
• Provide automated disaster recovery for complex stateful applications across all storage media types
• Secure user groups and resources
• Enable efficient use of edge resources, low footprint platforms and data schemes

As part of hosted MEC solutions, users will expect the ability to not only assert parameters and lifecycle management states, but also to monitor their solution. Some of these capabilities will likely be made by the MEC app’s element management system that typically operates one application or location at a time. However, this is not enough. Users will want additional information providing deep insight into all elements: physical resources, cloud platform, NFs and services. Furthermore, they will wish to correlate and display views across any strata – from a full drill-down to a solution-wide view, including NF, application, service, pod, node, cluster and server, across multiple clusters and physical locations – as MEC solutions are highly distributed.

Great care must be taken in choosing a solution with interfaces that display not just a single application, but across the board information, making it easy to view the dependencies at the solution level, application level, platform level and resource level, while hiding other resources that belong to the operator and other tenants.

Robin Cloud Native Platform (CNP): A revolutionary cloud platform

Robin.io’s best-of-breed Kubernetes-based CNP combines 1-click application onboarding with declarative, context- aware workload placement, pinning your NFs and services to automated policies. Just tell CNP what resources and supporting applications your service needs, and it will auto-discover and configure them for you, as per your policy, over the entire automated lifecycle of the service. Add, stop, start, heal and migrate with ease.

With CNP, resources are modeled on numerous NUMA-aware options including memory, CPU cores, HugePages, overlay/ underlay networks and redundancy, applying affinity and anti-affinity rules as needed.  This also extends into the compute and storage placement and locality.

Robin Cloud Native Storage (CNS): Application- and Kubernetes-aware storage

The highest performing storage for Kubernetes, Robin CNS brings advanced data management capabilities to Kubernetes. It’s 2-3x faster than the nearest competition, and is capable of delivering bare metal performance across a range of workloads.

Robin Multi Data Center Automation Platform (MDCAP): Orchestration, automation and lifecycle management

Robin MDCAP orchestrates and manages the lifecycle of any workflow including, bare-metal provisioning, cloud platform instantiation, network functions (NF) lifecycle management, network services (NS) lifecycle management and Methods Of Procedures (MOPs). All of these can be triggered through a policy engine. MDCAP’s automated workflows support cloud-native network functions (CNF), virtual network functions (VNF) and 3rd party physical network functions (PNF), simultaneously.  All of this comes with a full stack observability suite and planning tools.

Robin MDCAP’s provides intuitive context-aware lifecycle management for your NFs, services, 3rd party applications and Kubernetes cloud platform, but it integrates those workflows with your physical platforms, like bare-metal servers and third-party appliances.

The Robin.io advantage

Robin solves all of the major edge cloud concerns, unifying edge to core with the following capabilities:

• Ease of management:
  • Manage 10,000s of edge nodes with one-click automation
  • Move from test to secure deployment rapidly and easily
  • DevOps – CI/CD/CT integration
• Full solution observability:
  • Bare metal, clusters, applications, services
  • Event correlation
• Low footprint:
  • Edge cloud platform that scales up and down
  • Cloud-native, K8s aware, storage services and data
• Container and VM support:
  • No operations or resource silos
  • One system, same operations tools and paradigms
• Automated DR for complex, stateful, edge applications
• Multitenancy and granular roles based access
• Policy driven design – not manual
  • Allow the user to easily define policies that eliminates hunting and hardcoding
  • Multi-domain policy control that correlates things we don’t regularly associate with cloud platforms, switches, SDx controllers, and transport
  • Network slicing for 5G use cases (cloud and transport)
• Secure operation
  • Block access between namespaces
  • Isolate resources
  • Enforce usage quotas
  • Restrict networking access

For more information on our MEC solution advantages and other 5G related automation and orchestration solutions, check us out at www.robin.io.