Network Technologies for Big Data Applications

The massive deployment of the Internet of Things (IoT) concept will lead, in the next decade, to a growth of two orders of magnitude in terms of the amount of data to be collected and analysed over the telecommunication networks. Ericsson Hungary is exploring high performing and reliable big data frameworks taking advantage of distributed (in geographical sense) systems for applications that will characterise the everyday life of society in the near future. The EIT Digital Doctoral School announces thus an open position for an industrial doctorate in Budapest on the specific topic of networking technologies underlying of data analytics frameworks. Concepts will be prototyped and their operation will be analysed in real use case scenarios. A 6-month mobility at KTH in Sweden, one of the best European research institutions in this field, is also part of the program.

Challenge

Today big data analytics software can handle live data streams quickly and efficiently, the latter in terms of available computing resource utilization and process scheduling. However, these solutions typically do not consider the performance of the underlying network that may become a bottleneck. With the help of Software-Defined Networking and Network Function Virtualization, data analytics functions can be virtualized: this leads to extra computing overhead but offers the possibility to tackle the performance issues of the underlying networks. Furthermore, with such technologies, we have the opportunity to place virtualized big data Virtual Network Functions (VNFs) as close as possible to the data sources decreasing the bandwidth load of the network. This approach can provide higher reliability to the deployed service by monitoring, scaling, healing and moving around VNFs in the virtualized platform. In order to apply these functionalities we need to use stateless VNFs, which means that VNFs and industrial IoT systems can externalize their states to multiple low-latency shared memory systems. There are numerous questions related to big data analytics frameworks that have to be solved in order to apply such approaches and to deploy analytics services optimally in a geographically scattered infrastructure.

Approach

The investigation areas can be grouped into four phases that must be completed during the PhD program. The first phase is the clarification of current and future use cases; second, the applicant models analyses, designs, evaluates, and implements prototypes of the proposed new concepts of big data analytics infrastructure; in the third phase, the proposed systems are studied through comprehensive simulations and measurements in real-life deployments; finally, leveraging on the results, the proposed solutions are to be further improved for a few well-specified use cases to reach the highest achievable performance and efficiency in the process of the data analytics service deployments in those areas.

Expected outcome

The expected results of the PhD are data analytics control solutions that enable live analysis, adaptive management and high reliability of the deployed services. The results shall include: comprehensive study of the role of the networks in terms of orchestrating and deploying big data analytics system components; detailed analysis of the impact of network performance on data analytics systems; a prototype of big data stream analytics system for a virtual platform deployed on a distributed infrastructure testbed in real life environment in cooperation with Ericsson; published papers in high-quality academic conferences and journals; implementations that are usable in core and 5G Telco products of Ericsson.

Location

The doctoral student involved in this program will share its time between the Co-Location Center of the EIT Digital Budapest Node, the premises of Ericsson Hungary, and the Budapest university of Technology and Economics. A 6-month mobility to KTH in Sweden, one of the best known European research groups in this field, will be also part of the program.

Facts

  • Industrial partner: Ericsson Hungary Ltd.
  • Academic/research partner: Budapest University of Technology and Economics
  • Number of available PhD positions: 1 
  • Duration: 4 years
  • This PhD will be funded by EIT Digital, Budapest University of Technology and Economics, and Ericsson Hungary Ltd.

Apply

Those interested in applying should send an e-mail to zoltan.istenes@eitdigital.eu, including a CV, a motivation letter, and documents showing their academic track records. Please apply before April 20, 2018.

READ HERE THE FULL POSITION PAPER