Doctoral INPhINIT Fellowships Programme - Next Generation IoT Service Placement Optimization with Fog/Edge computing
Type of Position: PhD Scholarship (Bolsa de Doutoramento)
Type of Contract: Research grant
Duration: 36 Months
Limit to reply: 2020-Feb-04
email@example.com RESEARCH PROJECT/RESEARCH GROUP
Website of the Computing Systems and Communication Networks, which encompasses the Communication Networks group.
POSITION DESCRIPTION -Research Project / Research Group Description:
The Next Generation IoT (NGIoT) is the evolution of the Internet of Things (IoT) to become more secure, safe, trusted and human centric. NGIoT will become offer advanced and tactile IoT interfaces to the end user, together with hyperconnectivity, edge computing, distributed Ledger Technologies (DLTs) and Artificial Intelligence (AI) to optimize information processing.
Fifth generation (5G) is one of the main enablers, providing high capacity mobile communication to NGIoT edge components. Transmission speeds in the order of the Gbit/s and submillisecond latency will enabling resource heavy applications, intelligent distributed camera networks and autonomous vehicles. However, the current cloud based computation and data delivery model does not allow the required quality of service (QoS) guarantees to be efficiently harnessed, due to the number of hops of wired networks between the 5G base stations and the cloud, which leads to a significant increase in latency.
Moreover, in a massively sensed world, forwarding all the data generated by devices directly to the cloud may consume the available bandwidth and lead to congestion. Therefore, it is necessary that processing be hosted near the devices, close to the source of the data, so that the high speed transmission of 5G can be utilized and data can be processed and filtered out by the time it reaches the cloud.
This bringing down of computation, storage and networking services to the network edge opens up many new research areas of applying Fog Computing over the 5G cellular network architecture. One of these research areas is service placement and migration optimization, whereby computing resources are assigned to service instance modules in the edge, fog and cloud in such a way as to meet application requirements and maximize resource management efficiency.
-Job position description:
This task will be hosted by the Communication Networks group of INESC-ID. The task consists in performing innovative research on the fog placement and migration optimization schemes in the context of NGIoT. Road safety in autonomous vehicle scenarios is regarded one of the most promising and challenging applications, though other IoT application may also be addressed. Real-time requirements, coupled with the need of communicating high data rates, clearly demand edge/fog computing solutions.
The accepted fellow will depart from existing related work on fog/cloud resource management and research innovative schemes that surpass or significantly improve the latter. Iterative optimization methods and genetic algorithms have been frequently used in related work, including previous work by our group.
In order to become more scalable, two additional techniques will be investigated and possibly integrated: Hierarchical partition of edge/fog resources and with algorithm instances running in parallel at the same and different levels. Optimization based on Machine Learning techniques. This consists of using tools such as deep neural networks to capture the characteristics of efficient resource allocation patterns, in order to spawn similar solutions in in similar situations. This technique should provide faster responses, once the neural network is trained. Our group has expertise and previous experience on 5G resource management, IoT and fog computing performance modelling. Regarding simulation tools, our group has already developed significant extensions to iFogSim , a state-of-the-art simulation tool relating IoT, cloud, fog and edge computing. However, this does not preclude the analysis of other simulators regarding their suitability to support this particular project.
The accepted fellow is expected to enroll in the PhD Programme in Electrical and Computer Engineering or the PhD Programme in Computer Science and Engineering at Instituto Superior Técnico – Universidade de Lisboa.
António Manuel Raminhos Cordeiro Grilo