Table 1 Summary of research gaps

[21]

Diego

Heuristic algorithms

Execution time

The prototypes described were extending to wider environment; integrated into planned cloud services.

[22]

Multiopt

Virtual machine

Response time

To move containers without affecting or reducing the use of cloud services.

[23]

MOO-ACA

GA_MOCA algorithm

Network transmission overhead

Use scheduling methods in cloud containers to reduce the problem of algorithm time.

[24]

EECS

APSO

Temperature

Create a cloud environment for IoT applications that is dynamic and container-based, and allocate apps to the most appropriate containers.

[25]

Container-based virtualized model

VM

Execution time

Analyze the impact of post-failure work restructuring, interruptions due to work proximity in multiple cloud environments

[26]

Adaptive fair-share method

GPU memory allocation algorithm

GPU memory utilization

Improved Tensor Flow multi-container processing allows to securely share a GPU

[27]

ECSched

MCFP algorithm

Fraction of containers

To embrace more intricate circumstances, consider container dependencies and resource dynamics in the scheduler.

[28]

SRPSM

VM

Sensitivity

Searching multiple containers on same VM to perform multiple tasks in parallel

[29]

KCSS

Machine learning

Computing time

KCSS to identify residential containers and improve global performance.

[30]

CANSS

Naive Bayes

Cache hit ratio

Use artificial intelligence algorithms to compute if cache localization can be achieved

[31]

State-of-the art scheduling algorithm

Optimization algorithm

Throughput

Create a security alert table to avoid security issues related to the use of containers in your cloud infrastructure.

[32]

Skippy scheduling container

MCDM algorithm

function execution time

By implementing high-level operational goals, customize key planning parameters to explore specific aspects.