Table 3.

Key points of the analyzed articles in the data replication and placement techniques in edge computing

Source	Main idea	Technique	Results
Tziritas et al. [73]	Examining data replication and virtual machine migrations to mitigate network overhead in edge computing systems	Proposing an algorithm based on hyper-graph partitioning	The algorithm yields up to 53% network overhead reduction than state-of-the-art algorithms found in the literature
Aral and Ovatman [67]	Proposing a decentralized replica placement algorithm for edge computing	A method for disseminating data depends on the dynamic generation, replacement, and removal of replicas, led by constant monitoring of data requests originating from the underlying network's edge nodes	When opposed to client-side caching, which is frequently utilized in conventional distributed systems, a decentralized replica placement technique offers considerable cost savings
Li et al. [30]	Proposing replica creation and selection strategy based on the edge cloud architecture	Suggesting the DRC-AH (Dynamic Replica Creation Algorithm based on access Heat) and replica selection algorithm -NSC (Node Service Capability) algorithms (DRS-NSC)	The suggested algorithms have substantial benefits in prediction accuracy, resource usage, and other areas, user request response time, enhancing the system's performance to a degree
Li et al. [39]	Proposing flexible replica placement for enhancing the availability in the edge computing environment	The replica placement algorithm (FNSG) and (DRC-GM)	In an edge computing context, the DRC-GM and replica placement-FNSG algorithms may significantly enhance system performance regarding prediction accuracy, effective network and storage space use, access response time, increasing data availability
Bhalaji [11]	Examining efficient and secure data utilization in mobile edge computing by data replication	Proposing method was simulated using the network simulator-2	The duplicating method improves bandwidth usage while lowering power usage and reaction time
Saranya et al. [35]	Proposing data replication in mobile edge computing systems to reduce latency in IoT	Proposing both simple and random algorithms of replication	When compared to a straightforward replication method, a random technique for replicating can reach a bandwidth that is superior in terms of savings
Wang et al. [36]	Adaptive data replication in real-time reliable edge computing for IoT	Presenting the adaptive real-time reliable edge computing architecture	The suggested data replication techniques and architecture can provide the required levels of data loss tolerance while reducing network bandwidth use and preserving latency performance
Li et al. [77]	Examining effective replica management for improving reliability and availability in an edge-cloud computing environment	Proposing a dynamic replica creation strategy based on the gray Markov chain	While guaranteeing load balancing, the suggested dynamic replica generation approach significantly decreases system response time, increases data read throughput, and enhances system storage space usage
Li et al. [69]	Proposing resource and replica management strategy for optimizing financial cost and user experience in edge cloud computing systems	Proposing a dynamic replica allocation strategy and the replica consistency preservation strategy	The suggested resource management method may lower the overall financial cost of the rented nodes and the SLA (Service Level Agreement) default rate while also improving CPU (Central Processing Unit) usage as time goes on
Tang et al. [50]	Proposing a new replica placement mechanism for mobile media streaming in edge computing	Implementing the greedy algorithm of the mechanism	The technique can minimize the customer's access time, the replica availability, and the server's load balancing