Designing an Edge-Cloud Platform for Ultra-Low Latency and Data Intensive Applications of the Future: 5G Services and XR Applications

Edge-cloud and mobile edge computing are emerging concepts extending traditional cloud computing by deploying compute resources closer to the end devices. This approach, closely integrated with carrier-networks, enables several future services, such as 5G and beyond telco services and augmented reality (AR)/virtual reality (VR)/mixed reality(MR) applications. The emphasis is on integration: the rigorous delay constraints, the dynamicity, the reliability and the sheer scale altogether call for a high performance platform that offers the applications a simple but powerful API to the underlying compute and network infrastructure. The goal of the PhD thesis is to model, analyse, design and implement such a resource and service management layer. The major innovation outcomes are an edge-cloud platform prototype, and patents on the resource management methods applied within.


Today’s management algorithms of virtualized infrastructure resources are usually designed for a single data centre; therefore, they do not consider the characteristics provided by the underlying network, which can be a bottleneck when the compute infrastructure contains remote edge nodes. While the number of connected mobile devices and the amount of data to process grow, the efficiency of resource use gets more important. Furthermore, future use cases of 5G networks, e.g., MR applications and Tactile Internet, dictate strict latency and jitter requirements that have never been seen before. These phenomena call for advanced resource management for these future applications that must be run on a hybrid edge-cloud infrastructure. The first and foremost challenge of edge computing is the optimal scheduling of virtualized functions with wide-area networking, which greatly affects the performance of applications. Moreover, the resource scheduler must deal with the mobility of end users/devices and must work reliably at scale.


The intended solution to the challenges is a generic edge-cloud management platform that provides the features that the envisioned novel services and applications require. Innovation will be required in three aspects: i) resource management methods that take into account the available network resources for distributed applications; ii) a slick-design platform of edge technologies and concepts, which can work with multiple virtualization platforms (Docker, OpenStack, Amazon AWS, Microsoft Azure, etc.), iii) Artificial Intelligence (AI)-based scaling solutions in order to manage the dynamicity of customers and of their demand. The proposed work includes:

  • specifying use-cases for large scale distributed applications in the fields of traditional Telco, 5G services, and AR/VR/MR and big data applications;
  • studying the deficiencies of the existing virtualization platforms and service management algorithms, particularly regarding the network resources;
  • designing novel function placement/scheduling solutions;
  • addressing the challenges posed by mobility of customers and unreliability of remote infrastructure;
  • designing and implementing the concepts as platform prototypes, which consider the underlying network characteristics and support mobile agents as well.


The expected outcomes of the programme are the design elements and the evaluation of an edge-cloud resource management platform that is compatible with the major virtualization platforms and provides proper resource scheduling for distributed applications with strict latency requirement and/or producing data-intensive network load.

While one of the PhD students will focus on time-critical applications (e.g., tactile Internet) and the requirements they pose on the platform, the other PhD student will focus on data-intensive applications (e.g., big data analytics on IoT-generated data streams) and the related caching and pre-processing elements of the platform.

Moreover, the results shall include in both cases:

  • A comprehensive study of the available edge-cloud resource orchestrators and their network-aware capabilities;
  • Analytic models of the interplay between the edge-cloud infrastructure and the future applications;
  • A newly designed resource scheduling algorithm, a user location prediction algorithm, and a service management algorithm for various dynamic scenarios and application workflows;
  • Evaluation of the algorithms: both analytical (e.g., complexity analysis) and numerical (e.g., optimality and approximation), formal proofs on the efficiency and complexity of the proposed algorithms;
  • A verified prototype implementation of the edge-cloud resource and service management platform;
  • Published papers in high-quality academic conferences and journals.
  • Demos of the platform with selected application prototypes at academic events with high industrial interest;
  • Patents on the proposed platform with Ericsson;
  • Prototype implementations that are re-usable in 5G Telco products of Ericsson.


As an industrial doctoral student, you will reside in the EIT Digital Doctoral Training Centre in Budapest and share your time with the premises of Ericsson Hungary and the Budapest University of Technology and Economics. A 6-month mobility to another European university or research institution will be also part of the programme.


  • Industrial partner: Ericsson Hungary Ltd.
  • Academic/research partner: Budapest University of Technology and Economics
  • Number of available PhD positions: 2
  • Duration: 4 years
  • This PhD will be funded by EIT Digital, Budapest University of Technology and Economics, and Ericsson Hungary Ltd.


If you are interested in applying, please send an e-mail to Rolland Vida ( including a CV, a motivation letter, and documents showing your academic track records.

Please apply before 25 July, 2019.


© 2010-2021 EIT Digital IVZW. All rights reserved. Legal notice