To understand the industry, technology and application evolutions driving edge computing, Heavy Reading conducted a survey in collaboration with Accedian, Kontron and Red Hat. The survey questioned 82 communications service providers (CSPs) that have launched edge computing solutions or are planning to do so within 24 months. Highlights of the survey are discussed in Heavy Reading's white paper, "Harnessing Edge Computing for 2022." One of the drills-downs of the survey is in the area of performance management and assurance: the top challenges, desired functionality and changes in who, or what, are the primary users of metric data and KPIs.
Assuring edge performance
4G/5G network interoperability ranks as the top challenge in assuring the performance of multi-access edge computing (MEC)-based services (see figure below). This is not too surprising since the majority of Heavy Reading's survey respondents represent mobile and converged operators with both 4G and 5G networks – the so-called brownfield operators. A perpetual priority among these operators is preserving capital investment by ensuring generational interoperability between networks. According to this survey, "4G/5G network interoperability" is even more important than "security at the edge." CSPs consistently rank security concerns first or second, one reason being that security must be designed into MEC solutions from the beginning. It cannot be an afterthought or a layer on top of an already operational MEC service. All remaining challenges have the potential to materially impact MEC performance: managing virtual service layers, containers and microservices, network slices and so on. So, how do the CSPs prioritize these remaining challenges?
It is interesting that the responses from CSPs planning to deploy edge computing are fairly homogeneous, with only a 30-percentage-point difference between their biggest and smallest challenges. CSPs that have already deployed edge computing, on the other hand, have a spread of 56 percentage points. The top three challenges are the same regardless of deployment status, though they are of far greater concern to those who have already deployed MEC.
The remaining challenges vary widely in order of importance between respondents with MEC in production versus those still in the planning stage. Real-world experience with the edge clearly informs these responses. Assuring service performance at the edge is complex, and many aspects of it must be built into the solution from its inception. Controlling and managing edge performance, gaining insight from KPIs, and correlating virtual and physical layers for automation all must be considered and built in at the service design phase. They cannot be layered on as an afterthought – just as any effective security solution must be designed in from the beginning.
CSPs with MEC services in production also show a heightened concern about the fourth and fifth challenges in the figure below, both of which invoke automation in terms of either the insights needed for automation or automated active network tests. These items are key to managing the lifecycle of a large and highly distributed edge service. Where the concerns of CSPs with deployed services fall off are with the following:
- Challenges 10 and 13: Both address network slicing (still in the design phase of most CSPs)
- Challenge 11: Highly dynamic services – as today's MEC services are not, by and large, highly dynamic in terms of location or workloads
- Challenge 12: Ad hoc (as opposed to automated) testing
Read the full report for more edge computing insights
Heavy Reading's survey results show that carriers have committed to edge computing and are progressing rapidly with implementations. The deployment of edge computing brings with it issues of scale and complexity. CSPs are most concerned with overall network performance and security. In fact, those companies that have already deployed the edge have a heightened concern about these issues. They are looking for help from their traditional vendor and integrator partners, from their network monitoring and assurance tools and from the hyperscalers. To gain more in-depth details on service providers' perspectives on edge computing deployment, download and read the full report now: Harnessing Edge Computing for 2022.
— Jennifer Clark, Principal Analyst, Cloud and Infrastructure, Heavy Reading
This blog is sponsored by Accedian.