Siemens AG (Munich) is searching for Interns to work on the Internet-of-Things topic. If you are interested, please check here.
A Packet-level Simulation Environment for Edge Computing
- Creating edge computing scenarios with network simulator NS3
- Tailoring cloud simulators for mobile/wearable devices, edge devices, and the distant cloud
- Bridging network simulator and cloud simulator to create a complete edge computing simulation environment
EdgeOS: A Micro-service-based Edge Computing Framework (Ongoing)
- OpenWhisk-based backend service running in Docker
- Live migration of Docker to handle user mobility or handover between networks
- Andriod APP front end
Dynamic Configuration of IoT Environments (Ongoing)
Jointly with Siemens AG.
The Internet of Things (IoT) is growing at rapid speed. In the near future, the Internet of connected “things” will largely outnumber the humans who are currently using the Web. Industry estimates show that 30-50 Billion things will be connected to the Internet by 20201. For companies such as Siemens, the IoT concept plays a tremendous role for their ongoing evolution towards full digitalization, since possible applications range from smarter cities and eHealth systems, to Industry 4.0. Another example of such a business application for IoT are building automation systems, which will transform into IoT environments, where the various components of a building infrastructure (e.g., lights, switches, shutters, AC, controllers, etc.) will be IoT-enabled. Hence, in this thesis, building automation systems will be taken as a use case example.
Inter and Intra DC Orchestration of Network Services (Ongoing)
Jointly with Universität Göttingen.
Benchmarking System for Cloud-based, Distributed Machine Learning Applications (Finished)
Khurram Memood Cheema, jointly with AGT International.
The thesis will focus on design and implementation of a benchmarking platform for cloud-based machine learning applications.The benchmarking will be done by the platform in a systematic and automated way. The benchmarking system will be based on the HOBBIT benchmarking platform which is currently being developed as a part of HOBBIT EU project. The purpose of the project is to provide open benchmarking reports which can help EU industry to adopt Big Linked Data solutions. The contribution will extend HOBBIT to benchmark cloud-based application which is not the case at the moment.