Our previous blog post talks about how Edge computing will evolve in the context of the 5G revolution and discusses its applications across industries. In this blog, we will explore the Multi-access Edge Computing (MEC) framework and its primary components.
But before we do the heavy lifting on MEC, let's understand the basics.
MEC: Overview
MEC moves the computing of traffic and services from a centralized cloud to the edge of the network and closer to the customer. Instead of sending all the data to a cloud for processing, the network edge analyzes, processes, and stores it. Collecting and processing data closer to the customer reduces latency and brings real-time performance to high-bandwidth applications.
MEC characteristics include:
- Proximity
- Ultra-low latency
- High bandwidth
- Virtualization
MEC also offers cloud-computing capabilities and an IT service environment at the edge of the network. One typically implements MEC with data centers distributed at the edge, where applications require high bandwidth and a low latency environment. To achieve that, service providers create distributed data centers or distributed clouds. The resources that make up a cloud can reside anywhere—from a centralized data center to a cell site, a central office, an aggregation site, a metro data center, or the customer premises. The MEC platform enables distributed edge computing by processing content at the edge using either a server or a CPE.
A software-defined access layer can also be an extension of a distributed cloud. Most edge computing initiatives use open-source hardware and software that leverage cloud and virtualization paradigms, including SDN and NFV.
Benefits of MEC
Telco edge computing refers to distributed computing, managed by the CSP, extending beyond the network edge and onto the customer edge. Customers can run low latency applications and cache or process data close to the data source to reduce backhaul traffic volumes and costs.
Telco edge cloud incorporates the benefits of cloud and local computing, and 'local' can be on-device or on-premises. As a result, customers can run low latency applications and cache or process data close to the data source. Doing this also reduces backhaul traffic volumes and costs.
There are multiple potential locations for telco edge computing on and off the public network. These include customer premises, cell towers, street cabinets, and network aggregation points in the access and core network.
The decision for where to incorporate edge compute infrastructure for a telco depends on three factors:
- The telco's current network architecture,
- The virtualization roadmap, which is where you plan data center facilities for network applications, and
- The demand and use cases telcos cater to.
For example, cell towers are close to the customer edge and cover a wider area than street cabinets. These would suit low latency communication for autonomous vehicles – allowing real-time responses for the cars based on the current environment and happenings further up the road.
Understanding the MEC Framework
ETSI coined the MEC framework to standardize MEC deployment in a 3GPP compliant CSP ecosystem. The key USPs of this framework are represented below.
MEC enables the implementation of MEC applications as software-only entities that run on top of a virtualization infrastructure located in or close to the network edge. The MEC framework shows the general entities involved.
These entities can be broken into system level, host level and network-level entities. The figure illustrates the framework for Multi-access Edge Computing consisting of the following entities:
- MEC host, including the following:
- MEC platform;
- MEC applications;
- Virtualisation infrastructure;
- MEC system-level management;
- MEC host-level management;
- External entities or network-level entities.
MEC Applications
The key functionalities of these applications are:
- These applications run as virtual machines
- They typically consume but may also offer mobile edge services
- Apps provide information to aid the lifecycle management (e.g. indication of availability)
- The apps have associated rules and requirements and rules regarding:
- DNS configuration and traffic redirection (originate/inspect/modify)
- Compute/Storage/Networking resources, maximum latency, required services.
The rules and requirements are validated by mobile edge system-level management
- The Edge apps are assisted by mobility information may relocate user state
- If supported, the apps may be migrated to another mobile edge host
MEC Platform
- It provides an environment where applications can discover, advertise, consume, and offer mobile edge services
- It acts as a controls-plane in the virtualization infrastructure ("SDN") based on traffic rules
- DNS proxy/server configuration is done based on DNS records from the mobile edge platform manager
- It provides mobile edge services and may also consume mobile edge services offered by the applications
- It provides access to persistent storage and time of day information
MEC Orchestrator
MEC Orchestrator maintains an overall view of the system and mobile edge hosts, available resources, available services and topology and on-boards application packages, including:
- Integrity check and authenticity
- Validation of the applicable rules and requirements, and if necessary, adjust them to comply with operator policies
- Maintenance of a record of on-boarded packages
- Preparation of the virtualization infrastructure manager(s) to handle the apps
- Selection of the appropriate host for the application, satisfying its rules and requirements
MEC Orchestrator triggers application instantiation and termination and optionally triggers application relocation.
MEC System Level Management: OSS
This includes:
- Providing the network operator to trigger management and control actions, including the configuration of policy.
- Receiving requests from external entities (User app LCM proxy, CFS portal) for mobile edge application instantiation or termination.
- Receiving requests from the user app LCM proxy for mobile edge applications relocation between external clouds and the mobile edge system (optional).
- Determining if requests can be granted and forwarding granted requests to the orchestrator.
MEC Platform Manager
The functions of the MEC platform Manager include:
- Element management of the mobile edge platform.
- Lifecycle management of mobile edge apps involves notifying the orchestrator of app-related lifecycle events.
- Management of application rules, including service authorization, traffic rules, DNS configuration, conflict resolution.
- Receives virtual resource fault reports and performance measurements from the VIM and processes them.
Conclusion
At Quantiphi, we build our solutions leveraging this framework and create our containerized applications as microservices in the MEC application layer. We use multiple cloud management platforms from hyper scalars for managing multi-cloud ecosystems, which act as MEC orchestrator and MEC platform manager. Our solutions leverage the 5G MEC ecosystem to create seamless experiences for our partners, the communication service providers, and the enterprises. We build horizontal applications across industries with our intelligent video analytics and edge AI solutions.
Get in touch with our experts to explore our MEC capabilities.
Read more on "Quantiphi's Partnership with Google Cloud to deploy intelligent video analytics and low-latency Edge AI solutions."
