Skip to Main content Skip to Navigation

Proximity-Aware Replicas Management in Geo-Distributed Fog Computing Platforms

Abstract : Geo-distributed fog computing architectures provide users with resources reachable within low latency. However, fully exploiting the fog architecture requires a similar distribution of the application by the means of replication. As a result, fog application replica management should implement proximity-aware algorithms to handle different levels of resource management. In this thesis, we addressed this problem over three contributions. First, we designed a proximity-aware user-to-replica routing mechanism. Second, we proposed dynamic tail-latency-aware replica placement algorithms. Finally, we developed autoscaling algorithms to dynamically scale the application resources according to the non-stationary workload experienced by fog platforms.
Complete list of metadata
Contributor : Guillaume Pierre Connect in order to contact the contributor
Submitted on : Friday, December 18, 2020 - 4:42:09 AM
Last modification on : Wednesday, November 3, 2021 - 6:30:04 AM
Long-term archiving on: : Friday, March 19, 2021 - 6:24:57 PM


Files produced by the author(s)


  • HAL Id : tel-03080983, version 1


Ali Fahs. Proximity-Aware Replicas Management in Geo-Distributed Fog Computing Platforms. Operating Systems [cs.OS]. Université de Rennes 1, 2020. English. ⟨tel-03080983⟩



Les métriques sont temporairement indisponibles