|
| References | Resource provisioning techniques used | Major contribution | Pros | Cons |
|
| Zhang et al. [49] | Combined genetic algorithm and simulated annealing algorithm-based method | Mobility-aware cloud service allocation framework | Allocation decision was made using the power consumed by the mobile device when connecting to the cloud | Time delay for allocating resources not considered, thus leading to SLA violation |
| Manukumar [50] | Agent-based offloading decision maker | The decision maker chooses the compute component that runs on the cloud and mobile sides | Adaptable for bandwidth fluctuation applications | The context of the mobile requests has not been considered |
| Lee et al. [51] | Computer-intensive tasks to speed up execution and improve performance are utilized | Code-offloading architecture for native applications in the MCC environment | Native offloader improves execution speed | Optimized cross-platform translation of SIMD instructions is necessary for native offloading frameworks for multimedia applications |
| Ascigil et al. [52] | Function-as-a-Service (FaaS) for resource allocation | Application providers install their latency-critical processes that handle user requests with constrained turnaround times | Delay tolerance threshold of the user considered for offloading | The decision to offload is considered only based on power consumption |
| Nawrocki et al. [53] | Context-aware resource allocation using supervised learning agent architecture and service selection algorithm | Minimizes the cost while meeting mobile client requests’ deadlines | Cost-efficient scheduling based on energy context and Internet connection type | Impact of device mobility not analyzed |
| Farahbakhsh et al. [54] | Context-aware resource allocation | The monitor-analysis-plan-execution (MAPE) cycle is used to collect and analyze the circumstances before making decisions on offloading | Faster provisioning of dynamic resources | Does not provide better QoS for mobile customers due to the dynamic environmental changes |
| Chase and Niyato [55] | Optimal solution using deterministic equivalent formulations | Resource provisioning technique with joint optimization of VM and bandwidth reservation | Cost-efficient | Random network delays and VM migrations deviate the pricing |
| Mireslami et al. [56] | Branch-and-bound technique to obtain realistic discrete solutions | Cloud resource allocation problem with concurrent cost and QoS optimization | Provides optimized resource allocation | Handles workloads from only the web |
| Midya et al. [57] | Hybrid particle swarm optimization | A three-tier architecture made up of the local cloudlet, the centralized cloud, and the mobile cloud | QoS achieved | Slow convergence time |
| Chunlin and Layuan [58] | Lagrangian multiplier method | Multiple context-based service scheduling | More parameters are considered and thus there is improvement in mobile user’s QoS experiences | Impact of device mobility not analyzed |
| Quian and Andresen [59] | Offloading computations to resourceful servers | Energy-aware computation offloading is supported by the Jade runtime engine for smartphone platforms | Improves the performance of mobile applications and lowers energy usage | Compatibility issues for non-Android users |
| Niu et al. [60] | Bandwidth partitioning scheme based on weighted object relation graphs (WORGs) | Bandwidth-adaptive application partitioning algorithms and optimization models | Reduced energy use and enhanced performance with bandwidth adjustability | Requires an optimal solution for large-scale real-time applications |
| Zhou et al. [61] | Deadline-based resource provisioning | Mobile cloud offloading framework considering the user and cloud contexts | Reduced energy consumption for migrating apps | Applicability is for specific applications and single user |
| Durga et al. [62] | Context-aware resource allocation | Optimized allocation of resources based on cuckoo optimization | Minimized cost while meeting mobile client requests’ deadlines | Device mobility is not considered |
| Naqvi et al. [63] | Context-aware and cloud-based resource allocation | Visual augmented reality techniques | Lower latency and reduced memory load | Cost benefits yet to be analyzed |
| Naha et al. [64] | Dynamic context resource allocation | Resource ranking based on the constraints for the fog-cloud environment | Reduced processing time and cost | Failure handling to be included |
| Durga et al. [65] | Optimal resource allocation for balancing the costs and benefits of mobile users and cloud servers | Cuckoo search-based optimization algorithm and a context-aware cloud resource management algorithm | Improved QoS | Fog and edge devices can be utilized for load balancing |
| Spatharakis et al. [66] | Resource profiling mechanism | Two-level edge computing architecture with location estimation technique and scaling and allocating resource techniques | Increased performance by considering resource under-utilization and QoS criteria | Accuracy to be improved and can be extended for time-specific applications |
| Jia et al. [67] | QoS-aware cloudlet load balancing | Fast heuristic algorithm and distributed genetic algorithm | Minimizes the maximum task response time | Increased execution time |
|
