Mobile Edge Computing brings together 5G network operators and hyperscalers to move compute and data processing out of the data center and onto the network edge
Mobile Edge Computing or Multi–access edge computing (MEC) reduces latency by bringing compute power out of the data center and closer to the end user — on the edge of a network, instead of its center. 5G MEC enables a host of new business opportunities for network operators and enterprises, opening up opportunities for low-latency communications, massive bandwidth, better network traffic management, network slicing and massive IOT.
But redistributing computing and adjusting networking architectures to accommodate MEC presents challenges, both to network operators and hyperscalers. What’s more, MEC is still an emerging technology, with new business models and new technology constantly driving innovation. Right now, telecoms and hyperscalers are collaborating to bring new MEC solutions to light as enterprises increasingly find ways to use 5G and MEC.
The overarching principles that guide MEC make networks faster and more responsive to the needs of individual customers. MEC focuses resources on the network where they need to be to make the most impact. These ideas are not new to IT, or to the Internet. They’ve driven Content Delivery Networks (CDNs) and the distribution of software on the Internet for decades. Let’s look at some examples of how MEC computing infrastructure is deployed.
Google’s global MEC plans
Google Cloud in 2020 first articulated its Global Mobile Edge Cloud strategy. The company called GMEC its “portfolio and marketplace of 5G solutions built jointly with telecommunications companies.” GMEC leverages Google Cloud’s global WAN, comprising regions, 134 edge POPs and “thousands of edge nodes” in support of edge services.
GMEC is based in Anthos, Google’s fully managed hybrid cloud platform that works on premises, edge and in multiple public clouds, all through a Google Cloud-backed control plane. Anthos uses the Google-developed Kubernetes engine as the basis for its container orchestration management.
Anthos is built on Google Kubernetes Engine (GKE); Anthos Config Management, a policy and security automation engine; Anthos Service Mesh, which manages traffic and telemetry; and Anthos Migrate, which manages Virtual Machine (VM) to GKE migration. Also announced at Google Cloud Next were a preview launch of Anthos for Virtual Machines and Anthos Multi-Cloud API to help its customers manage Kubernetes clusters on AWS and Azure cloud environments.
Google’s vision for GMEC comprises a three pillar strategy: An edge solutions portfolio, an open platform and edge infrastructure spanning Google assets, the telco edge, customer edge and third-party edge. Google is working with Orange, Telefonica, Verizon and other prominent telcos to bring 5G MEC services online.
AWS Wavelength: MEC for 5G
Verizon’s 5G Edge service combines Verizon’s 5G Ultra Wideband network with AWS Wavelength. The service launched in Boston, Massachusetts and the Bay Area in 2020. Verizon and AWS later expanded it to 10 more cities. Verizon and AWS have also launched a private MEC solution to support industrial and enterprise applications unsuited for the public cloud.
AWS Wavelength is Amazon’s specialized 5G edge service. AWS Wavelength deploys compute and storage services on Amazon’s network into “Wavelength Zones” (WZs). WZs are managed by the control plan in the AWS Region where they operate. They’re a new type of AWS infrastructure designed especially to run workloads that require ultra-low latency over mobile networks.
Verizon began MEC deployment with its SAP (Service Access Point) points within each of its markets, which allows the carrier to deploy and manage edge computing capabilities to those sites. It’s also planning to expand into C-RAN locations. Vendors like AWS believe edge computing isn’t about technology but ecosystem, building a network of MEC-experienced developers who bring forth innovations and new solutions.
Microsoft Azure Edge MEC, with AT&T
Microsoft announced Azure Edge Zones in 2020 along with Azure Private Edge Zones. Azure Edge Zones connect directly to 5G networks; their private counterpart works with private LTE/5G network combined with on-prem Azure Stack Edge.
Microsoft’s global Azure Wide Area Network (WAN) comprises almost 200 points of presence over 60 regions, across 140 countries. What’s more, Microsoft has optimized traffic orchestration towards 5G requirements.
Collaborating with AT&T, Microsoft soft-launched the first Azure Edge Zone in Atlanta, Georgia in late November. The companies plan a Dallas Edge Zone next, with other plans for North American locations in 2022.
More broadly, Microsoft offers customers Azure Stack Edge, specialized appliances for edge computing applications which are rack-mountable. Azure Stack Edge can act as a cloud storage gateway and enables eyes-off data transfers to Azure, while retaining local access to files.
