Skip to main content

How can telcos use Mobile Edge Computing to leverage IoT scenarios?

(Image credit: Image source: Shutterstock/Chinnapong)

According to a recent European Commission study, the market value of the IoT in the EU is expected to exceed one trillion euros in 2020. How can global telcos ensure they get a good slice of the IoT cake and are not becoming pure bandwidth providers? Patrick Steiner, Associate Manager, Specialist Solution Architecture at Red Hat believes the answer lies in supporting Internet of Things scenarios with Mobile Edge Computing (MEC). Here, we ask him how MEC is impacting the IoT market and for his views on the future of the telco ecosystem.

Q: In a nutshell, what is Mobile Edge Computing (MEC)?

Mobile edge computing (MEC) is a highly distributed computing architecture based on the deployment of small computing units in both outdoor and indoor radio access network (RAN) facilities at the edge of the telcos’ networks. 

Q: What is the overall goal of using MEC?

The goal is to provide on-premises computing using standard x86 architecture rather than custom hardware that is able to run isolated from the core network. This will achieve a number of things: It will increase network agility, encourage innovation at the edge of networks, and also reduce capex and opex associated with spinning up new services.

Q: How do you define MEC use-cases, and how does it relate to IoT?

As a member of the European Telecommunications Standards Institute (ETSI), Red Hat is helping define MEC use cases and architecture details. We’re collaborating with the wider community to ensure common open standards prevail. Because MEC is isolated from the main network it is ideal for uses cases such as video analytics, location services, augmented reality, data caching, optimised local content distribution and, of course, IoT.

Q: Why should telcos choose MEC ahead of other architectures to provide new IoT driven service opportunities?

By using standard x86 architecture, along with a virtualised software stack, it is possible to decouple the software from the hardware. This speeds up time to deployment of new services, as only new software needs to be installed rather than new hardware. Telcos can also exploit economies of scale by consolidating many IoT applications onto the same industry-standard high-volume servers, switches, routers, and storage, transforming these environments into elastic, pooled resources that can scale up or down as needed. And, as you will see later, it also helps improve security.

Q: How can MEC help differentiate telco’s solutions from over-the-top (OTT) players? Can MEC even effectively host OTT platforms?

I’ll take the second part of the question first. In a word “yes” MEC can host OTT platforms. The key element here is the MEC IT application server, which is integrated at the RAN element. The MEC server platform consists of a hosting infrastructure (equivalent to the NFV infrastructure) and an application platform. In some ways, the MEC Application Platform is similar to an IoT gateway or a more classical IT Application Platform (like Java2EE).

It terms of helping telcos to differentiate their services, the key here is the wealth of data that service providers can leverage. With proximity services enabled and location awareness to capture and analyse key information (like geo-location and trajectory) from the user equipment, and thanks to very low-latency access to the radio channels and detailed network context information such as radio conditions and neighbours statistics the telcos will be in a position to offer data rich services that are beyond the scope of OTT offerings. Furthermore, the telcos’ solutions will be more efficient and cost effective.

Q: Where do you deploy MEC servers? 

The first release of the MEC working group focuses on a scenario where the MEC server is deployed either at the LTE macro base station (eNB) site, or at the 3G radio network controller (RNC) site, or at a multi-technology (3G/LTE) cell aggregation site, although wireless LAN (WLAN) deployments will be soon included. This should enable a common architecture for the first trials of IoT, LTE-U (LTE in unlicensed spectrum, 5G, cloud RAN (C-RAN), virtual content delivery network (vCDN,) mobile video delivery, and other distributed network function virtualisation (NFV) use-cases.

The MEC server platform offers a virtualisation manager (an infrastructure as a service or IaaS abstraction), and advanced services such as traffic offload function (TOF), Radio Network Information Services (RNIS), communication services and a service registry. This layer abstracts the details of the radio network elements, so the MEC applications are portable and compatible across the network thanks to the standard, open APIs that will be defined. In other words, MEC will define a platform-as-a-service (PaaS) equivalent, hosted in an NFVi IaaS such as OpenStack, with Openshift applications on top. These components and functional elements that are key enablers for MEC solutions in a multi-vendor environment. As enablers, they will stimulate innovation and facilitate global market growth, while leaving room for differentiation and value creation, in a similar way as the ETSI NFV has done since 2012.

Q: Is an open telco ecosystem, based on open standards and open APIs achievable? Will this be the future for telco companies? 

There is every reason to think that openness is the direction we’re heading. But there is an awful lot of legacy equipment locked into the system. One of the key benefits of open source is its potential to avoid vendor lock-in. The ETSI MEC concept introduces three management layers: one for the hosting infrastructure, like OpenStack, one for the application platform and one for the software applications and services. Using open source solutions like OpenStack for the infrastructure and JBoss for the application platform, most of the problem is now solved and the research can be focused on the definition of open APIs for the “northbound” interfaces with the software applications. At Red Hat we believe that combining ETSI NFV and EMS interfaces, with newly defined APIs like TOF, RNIS and the collaboration with NFV application development kits such as DPDK, OCP, or IOVisor, we can create an open telco ecosystem based on open standards and open APIs.

Q: Are there are security concerns?

Security will always be a consideration as part of any ICT implementation. Taking a decentralised approach to networking requires a new approach to security. However, that is far from being a bad thing. In October last year, a massive distributed denial-of-service (DDoS) attack occurred in the US that disrupted dozens of high profile websites, it happened because the connected IoT devices were communicating via centralised cloud services. By contrast, with an MEC model, such an attack would be contained within a local site, since critical functions are hosted at the edge itself, and edge devices do not need direct access to the internet. This makes it possible to systematically prevent such widespread problems from occurring in the first place. Further, this approach benefits users as well, as it lowers the response time of these services and therefore improves user experience.

Patrick Steiner, Associate Manager, Specialist Solution Architecture, Red Hat
Image source: Shutterstock/Chinnapong

Patrick Steiner
Patrick Steiner, Associate Manager, Specialist Solution Architecture at Red Hat is responsible for a team of 19 specialists located all over Europe.