Multi-access edge computing (MEC) provides cloud computing capabilities and IT services at the edge of the network. It introduces an intermediate computing element to offer advantages for applications running close to users. MEC aims to address challenges like improving response times, handling increased traffic from user mobility, and ensuring quality of service. Standardization groups like ETSI and 3GPP are working to fully specify MEC and its integration with 5G networks to support use cases involving augmented reality, intelligent video delivery, connected vehicles, and IoT gateways.

1. Introduction to
Multi-Access Edge Computing
March 2023
Evren Tuna

2. MEC - Close to the user, at the EDGE!
https://www.youtube.com/watch?v=crnPWql-0oo

3. MEC
• Offers to application developers and content providers cloud-computing capabilities and an IT service
environment at the edge of the network.
• Introduces a standard for supporting an emerging cloud paradigm for software development communities.
• Introduces an intermediate element at the network edge.
• Offers significant advantages to application components/services running there.
• Presents some challenges, e.g. higher cost, relatively small compute footprint, good local but not global
reachability, etc.

4. MEC
How do I get discovered
by my users?
How am I connected to the users?
How many users am I serving?
And where?
How to be sure I am running
when and where needed?
What if my users move?
How do I reach
my cloud service?
What is my QoS?
Where am I?
What is around me?
* https://www.etsi.org/images/files/ETSIWhitePapers/etsi_wp20ed2_MEC_SoftwareDevelopment.pdf

5. Standardization Groups in The Specification of Edge Computing

6. ETSI MEC
• Foundation for Edge Computing created
• Fully standardized solution to enable applications in
distributed cloud created by ETSI MEC + 3GPP
• Continuously growing MEC membership: 124
(updated Dec 2022);
• Diverse ecosystem: Operators – Technology Vendors
– IT players – Application developers - …

7. 3rd 3-year Phase of MEC Work Under Way

8. MEC toward 6G: planning MEC Phase 4
Transition from MEC Phase 3 to MEC Phase 4 can lead to:
• more consolidated work on MEC Federation, including exposure of resources managed by multiple
operators, e.g. addressing multi-domain and multi-tenancy slicing and MEC support for application slicing;
• MEC architectural/service updates needed to support cloud native communication systems and edge
native design for app developers (also with container support)
• introduction of proper normative work to improve security and privacy in MEC system.
• Further promotion of MEC as an attractive development environment for the industry by creating
“developer friendly environments” (e.g. portals, SDK, ..) that enable convergence of key industry
ecosystem, e.g., application developers and operators.
• Further outreach efforts, e.g. Hackathons/trials in collaboration with open-source communities, industry
groups (e.g. 5GAA, etc..)

9. MEC
• 3 “locations”: Client, Near Server, Far Server.
• MEC Host, usually deployed at the network edge,
contains a MEC platform and the compute, storage
and network resources for applications in VMs or
containers.
• The MEC platform offers a secure environment where
MEC applications may, via RESTful APIs, discover,
advertise, consume and offer services.

10. MEC Definitions
MEC Host
• Contains a MEC platform and a
virtualization infrastructure
• Provides compute, storage and
network resources to MEC host
MEC Application
• Can be instantiated on a
MEC host within the
MEC system
• Can provide or consume
MEC services
MEC Platform
• To run MEC applications
• To enable MEC applications
• To provide/consume services
• To provide MEC services
MEC Service
• Service provided via the MEC
platform either by the MEC
platform or by a MEC
application
1
2
3
4

11. MEC Application Start-Up
• MEC platform can verify the authenticity of the MEC
application with the aid of an AA entity that contains the
registration related information about the MEC
application.
• For actual authentication, MEC application uses access
token based on Oauth 2.0.
• MEC platform has received the valid configuration for
MEC applications (e.g. the required and the optional
services, the services to be offered by the application
instance and the associated transport dependency, the
associated traffic rules and DNS rules)

12. MEC Application Graceful Termination
• Upon receiving a request MEC platform notifies the MEC
application instance of graceful termination with a
timer.
• When the timer expires, MEC platform continues
terminating the MEC application instance.
• The MEC application instance has the option to inform
the MEC platform that it is ready to be terminated after
finishing application-level actions.

13. MEC Service Registration/Discovery
• When a MEC service is registered by the service
producing MEC application, the authorized relevant
applications (e.g. the applications that indicate the
service as “optional” or “required”) will be notified
about the newly available service.
• The authorized relevant applications will also be notified
about the service availability changes of that service.

14. Other Services Provided by MEC Platform

15. MEC Reference Architecture

16. MEC Reference Architecture

17. MEC Reference Architecture
On the left;
• Devices run application clients which either use the
DNS to discover application servers
• (Edge Application Server (EAS) in 3GPP SA6
terminology or MEC Application in ETSI ISG MEC
terminology)
• or use the Edge Enabler Client (EEC) to perform the
discovery according to the SA6 EDGEAPP architecture.

18. MEC Reference Architecture
Towards the middle;
• Platform (Edge Enabler Server (EES) in 3GPP SA6
and MEC Platform in ETSI ISG MEC) provide
functionality pertaining to mediating access to
network services, application authorization,
application's service registration and application's
service discovery, context transfer, etc.
• A given implementation can combine functions
specified by ETSI ISG MEC and ones specified by
3GPP SA6.
• The platform typically exposes APIs towards edge
cloud applications (MEC application or Edge
Application Server).
• EDGE-3 and Mp1 offer complementary API
functions, therefore can be considered to be part
of a single reference point from an application
developer perspective.

19. MEC Reference Architecture
Towards the right;
• Functionalities specified by ETSI ISG MEC include
management and orchestration of the MEC platforms
and OSS functions supporting access to portals offered
to application service providers.

20. MEC Phase – Study Item MEC in 5G (MEC 031)
• The ETSI white paper MEC in 5G networks sets
the scene for this study item
• ISG MEC investigates the opportunities offered
to MEC by the 5G system and its edge computing
enablers
• The scope includes the following
• C-plane interactions with 5GC,
• Functional split between MEC and 5GC wrt.
• API framework,
• Organization of MEC as an AF,
• Pertinent interactions of MEC with (R)AN

21. 3GPP enablers for MEC – Selection & re-location of UPF

22. 3GPP enablers for MEC – Selection & re-location of UPF

23. 3GPP enablers for MEC – AF influence on traffic routing

24. 3GPP enablers for MEC – Mobility event notifications

25. 3GPP enablers for MEC - Concurrent access to local and central DN

26. Use Case 1: Augmented Reality
• The processing of user location or camera view can be performed on a MEC platform rather than on a more centralized server.
• There may be a need to update information at a fast rate, depending on how the user moves, and the context in which the augmented reality service is
used (e.g. in an art gallery, exhibits are positioned only a few metres apart and each piece is supplemented with additional text on the artist, the
interpretation of the artwork, etc.)
• In other words, augmented reality data requires low latency and a high rate of data processing in order to provide the correct information to the user's
device depending on the location and orientation of the user.
• Performing such data processing on the MEC platform also has the advantage of collecting metrics, anonymized meta-data, etc., in order to analyse the
service usage and help to improve the service in order to provide a better user experience.

27. Use Case 2: Intelligent Video Acceleration
• In this scenario, a radio analytics application, which resides in a MEC server, provides the video server with an indication on the throughput estimated to
be available at the radio downlink interface.
• This information can be used to assist the TCP congestion control decisions (for example in selecting the initial window size, setting the value of the
congestion window during the congestion avoidance phase, and adjusting the size of the congestion window when the conditions on the "radio link"
deteriorate).
• The information can also be used to ensure that the application-level coding matches the estimated capacity at the radio downlink.

28. Use Case 3: Intelligent Video Acceleration
• Mobile Edge Computing can be used to extend the connected car cloud into the highly distributed
mobile base station environment, and enable data and applications to be housed close to the vehicles.
• This can reduce the round trip time of data and enable a layer of abstraction from both the core
network and applications provided over the internet.
• MEC applications can run on MEC servers which are deployed at the LTE base station site to provide
the roadside functionality.
• The MEC applications can receive local messages directly from the applications in the vehicles and the
roadside sensors, analyse them and then propagate (with extremely low latency) hazard warnings and
other latency-sensitive messages to other cars in the area (as depicted in Figure 5).
• This enables a nearby car to receive data in a matter of milliseconds, allowing the driver to
immediately react.

29. Use Case 4: Internet of Things Gateway
• Various devices are connected over different forms of connectivity, such as 3G, LTE, Wi-Fi or other radio technologies.
• In general the messages are small, encrypted and come in different forms of protocols.
• There is a need for a low latency aggregation point to manage the various protocols, distribution of messages and for the processing of analytics. The MEC
server provides the capability to resolve these challenges.
• Mobile Edge Computing can be used to connect and control devices remotely, analyse and provide real time provisioning and analytics.
• MEC enables the aggregation and distribution of IoT services into the highly distributed mobile base station environment, and enable applications to
respond in real-time.
• This can reduce the round trip time of data and enable a layer of abstraction from both the core network and applications in the cloud. IoT applications can
run on MEC servers which are deployed at the LTE base station site to provide this functionality.

30. References
• MEC federation: deployment considerations, ETSI White Paper No. 49, 1st edition – June 2022
• Harmonizing standards for edge computing - A synergized architecture leveraging ETSI ISG MEC and 3GPP
specifications, ETSI White Paper #36, 1st edition – July 2020
• ETSI MEC: An Introduction, ETSI MEC Leadership Team, Dec 2022
• Multi-access Edge Computing (MEC); Study on Inter-MEC systems and MEC-Cloud systems coordination,
ETSI GR MEC 035 V3.1.1 (2021-06)
• Multi-access Edge Computing (MEC); Framework and Reference Architecture, ETSI GS MEC 003 V3.1.1
(2022-03)
• Operator Platform Telco Edge Requirements, GSMA, Version 1.0, 29 June 2021

31. Thank You!
March 2023
Evren Tuna