BibTeX:

@article{IJIRSTV8I1017,
     title={A Study on Workflow Execution Time Prediction},
     author={Mohamed Ismail, Sajith Geo Paul, Thomas Mani, Rohith and Dr. Jyothish K. John},
     journal={International Journal for Innovative Research in Science & Technology},
     volume={8},
     number={1},
     pages={11--15},
     year={},
     url={http://www.ijirst.org/articles/IJIRSTV8I1017.pdf},
     publisher={IJIRST (International Journal for Innovative Research in Science & Technology)},
}

Several applications depend on the performance of workflow execution time prediction for differing input data. In any case, such estimates are hard to produce in the cloud. Cloud computing conditions give an expansive scope of focal points for the organization of logical application, especially the potential to provide a great number of resources under a pay-per-use plan. This trademark meets the needs of researchers who define applications as a collection of workflows. Scientific workflows consist of tasks with conditions among them. These workflows are huge in size and require a great number of resources to run. Some of the major drawbacks of the current model are the limited accuracy and the impractical pre-requisites required. The model we propose involves machine learning ensembles employing multiple learners capable of balancing out the weaknesses of each with the strengths offered by others. The proposed approach also uses extracted attributes as parameters to predict the execution time of workflow applications in dynamic environments such as cloud.

Cloud Computing, Runtime, Hardware, Machine Learning, Computational Modeling, Predictive Models, Analytical Models, Performance Prediction, Workflow Application Execution Time