|
References | Resource provisioning techniques used | Major contribution | Pros | Cons |
|
Shahidinejad and Ghobaei-arani [31] | DCloud: deadline-aware resource allocation algorithm | Dynamic adjustments in VM resource allocation | Dynamic resource allocation according to the deadline constraint | Allocating resources to mobile clients is insufficient |
Li et al. [32] | Energy-efficient deadline-based task scheduling | Compute-intensive jobs and non-compute-intensive tasks were scheduled in the cloud based on mobile device variables such as battery level and wireless connection type | Cuckoo search-based optimization algorithm was used to solve the NP-hard problem | The client’s mobility and the server’s current context have not been considered while evaluating the performance |
Durga et al. [33] | Prediction-based resource provisioning | Lightweight resource allocation framework | A profile created from a previous program run was used to predict the outcome | The CDC’s load and resource availability factors were not taken into account while predicting the execution time |
Chang et al. [35] | Level-based scheduling to find the suitable resources for cost savings and complete the execution within the deadline | Time and budget-aware scheduling algorithm for hybrid cloud | Cost optimization for resources allocated within the deadline | Other mobile client characteristics are not considered |
Praveen et al. [36] | Deadline and cost-based workflow scheduling | Cost optimization was done while allocating resources for a job | Level-based scheduling to find suitable resources for cost saving | The case of resource contention was not considered |
Tuli et al. [37] | Data-aware resource provisioning algorithm using Aneka | Deadline specifications for applications requiring a lot of data | The mean runtime of public cloud services is computed based on available bandwidth | Does not indicate how to estimate the execution duration of a job |
Nayak and Tripathy [30] | VIKOR-based task scheduling algorithm | VIKOR decision maker used to schedule similar tasks | Better resource utilization and reduces task rejection | Does not consider how to approximate response time |
Malawski et al. [42] | Resource provisioning and task scheduling algorithm based on resource utilization | It coordinates the scheduling of related jobs and settles disputes between them while allocating resources | (i) Allocates fixed resources initially and adjusts the number of resources to the need of applications (ii) Considers uniform resource distribution and budget in this approach | Deadline violation and job completion rate were not optimized |
Nayak and Tripathy [30] | Genetic algorithm-based workflow scheduling | Deadline and budget have been considered for making a scheduling decision | It achieves a lower cost while completing the task ahead of schedule | The problem’s single objective characteristic is the main flaw in their heuristic technique |
Alsadie et al. [39] | DTFA: a dynamic threshold-based fuzzy approach | Allows dynamic adjustments in VM launching time and bandwidth | Balances the cloud resource’s peak demand and thus reduces the request rejection rate | Insufficient to allocate resources to mobile customers |
Lu et al. [40] | Energy-efficient scheduling algorithm | Energy-efficient scheduling policy for cloud-assisted mobile computing applications | Offloads the code to achieve minimal energy consumption | Application deadline alone is considered for resource allocation decisions |
Shahidinejad et al. [17] | Novel autoscaling mechanism | Cost-efficient resource provisioning approach | Completes all the jobs within the deadline at minimum cost | Does not address the load conditions of the server |
Nadjaran Toosi et al. [43] | Data-aware resource provisioning algorithm | Data-intensive apps with user-defined deadlines | Data location, cloud startup time, network bandwidth, and data transfer time are considered | Does not consider job execution time |
Lai et al. [44] | Optimized stochastic user allocation (SUAC) algorithm | Optimized allocation of user’s multi-dimensional resource requirements | Effective and stable system for multi-criteria user requirements | Allocation for massive real-world requirements to be considered |
|