A Distributed Support based on Kubeflow to Ease MLOps of Federated Learning Algorithms

The main objective of my thesis work is to exploit the Google native and open-source platform Kubeflow, specifically using Kubeflow pipelines, to execute a Federated Learning scalable ML process in a 5G-like and simplified test architecture hosting a Kubernetes cluster and apply the largely adopted FedAVG algorithm and FedProx its optimization empowered by the ML platform ‘s abilities to ease the development and production cycle of this specific FL process. FL algorithms are more are and more promising and adopted both in Cloud application development and 5G communication enhancement through data coming from the monitoring of the underlying telco infrastructure and execution of training and data aggregation at edge nodes to optimize the global model of the algorithm ( that could be used for example for resource provisioning to reach an agreed QoS for the underlying network slice) and after a study and a research over the available papers and scientific articles related to FL with the help of the CTTC that suggests me to study and use Kubeflow to bear the algorithm we found out that this approach for the whole FL cycle deployment was not documented and may be interesting to investigate more in depth. This study may lead to prove the efficiency of the Kubeflow platform itself for this need of development of new FL algorithms that will support new Applications and especially test the FedAVG algorithm performances in a simulated client to cloud communication using a MNIST dataset for FL as benchmark.

Simulation of URLLC slicing with resource reconfiguration in 5G networks

This thesis seeks to analyse the performance of dynamic slice provisioning in a 5G metro network with the low latency and reliability guaranties. This elaborate highlight the comparison in terms of performance of two versions of a simulator developed in Python based on different models: the Exhaustive research model and Shortest Path First Fit (SPFF) model. It further presents the differences between the dedicated path protection and the shared path protection. This analysis is made through several simulations at different network conditions by varying networks resources and observing the network performances while comparing the 2 models mentioned above. A reconfiguration procedure was implemented on backup resources in the shortest path first fit in order to improve its performance with respect to the exhaustive research which is more optimised. Subsequently, several triggering events was implemented, for the reconfiguration. And a comparison is made between these different triggering events in terms blocking probability, bandwidth at link, capacity at each node, primary and backup bandwidth per slice and backup capacity per slice.