Optimization policy for file replica placement in fog domains

TL;DR

The paper tackles data availability and latency in fog computing by proposing a replica-aware data placement policy that uses complex-network metrics and graph partitioning to place three replicas per file. It combines community-aware network partitioning, Kernighan–Lin partitioning, and betweenness centrality to select replica locations, with a cloud-inclusive objective to balance proximity to sensors and to the cloud. Through 22 simulations and comparisons against a single-replica policy and FogStore, the approach achieves shorter read latencies, competitive file availability, and manageable network usage, albeit with higher write traffic due to multiple replicas. The work demonstrates a viable mechanism for improving data availability and latency in fog environments, highlighting practical trade-offs and paving the way for dynamic replica configurations and chunked-file extensions in future research.

Abstract

Fog computing architectures distribute computational and storage resources along the continuum from the cloud to things. Therefore, the execution of services or the storage of files can be closer to the users. The main objectives of fog computing domains are to reduce the user latency and the network usage. Availability is also an issue in fog architectures because the topology of the network does not guarantee redundant links between devices. Consequently, the definition of placement polices is a key challenge. We propose a placement policy for data replication to increase data availability that contrasts with other storage policies that only consider a single replica of the files. The system is modeled with complex weighted networks and topological features, such as centrality indices. Graph partition algorithms are evaluated to select the fog devices that store data replicas. Our approach is compared with two other placement policies: one that stores only one replica and FogStore, which also stores file replicas but uses a greedy approach (the shortest path). We analyze 22 experiments with simulations. The results show that our approach obtains the shortest latency times, mainly for writing operations, a smaller network usage increase, and a similar file availability to FogStore.

Figures (10)

• Figure 1: Example of a fog domain with storage capacities.
• Figure 2: Model representation of the example of Figure \ref{['fig:fogarchitecture']}.
• Figure 3: Example of the placement of the replicas of a file with two data sources.
• Figure 4: Results for the total latency for writing operations between the sensors and the file replicas.
• Figure 5: Results for the total latency for reading operations of the cloud provider in the file replicas.