Tools
Tools
Joy, Kripa Mariam (2022) DynamicForecast: Experts Council with optimized Workload Prediction Framework for Cloud Computing. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (879kB) | Preview |
Preview |
PDF (Configuration manual)
Download (2MB) | Preview |
Abstract
Resource management in cloud environment is a challenging task. Management of resources with predictive scaling technique has been used to circumvent the limitation of reactive scaling in the cloud such as over-provisioning and under-provisioning of resources. The predictive technique aids in workload (WL) predictors, which predict the fluctuations in workload. However, the accuracy of predictors varies as per the varied workload pattern. In order to address this, a novel workload prediction framework naming DynamicForecast (DF) has been introduced. DF works by using a stack of predictors with long-short-term memory (LSTM) model along with Adam's optimization technique to build the ensemble models which further predicts the workload more precisely. The performance is measured by Mean Absolute Percentage Error (MAPE). As MAPE decreases the accuracy of prediction increases. DF has 50% - 94% lower MAPE compared with state-of-the-art predictors (CloudInsight and ARIMA) for WLs which are bursty, random, and seasonal.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Cloud Computing; WL Prediction; Resource Management; Long Short-Term Memory; Machine Learning; Ensemble Model |
| Subjects: | Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Cloud Computing |
| Depositing User: | Tamara Malone |
| Date Deposited: | 08 Dec 2022 12:02 |
| Last Modified: | 08 Mar 2023 14:25 |
| URI: | https://norma.ncirl.ie/id/eprint/5983 |
Actions (login required)
 |
View Item |