International Journal Of Imaging

Abstract

"Leveraging Deep Learning for Optimized Workload Forecasting in Cloud Computing"

Author: Dr. FARHEEN SULTANA, Mrs. TAHERA ABID, Mrs. NAZIA AMREEN

Abstract:

Optimising cost-effectiveness and resolving problems like resource provisioning and Service Level Agreement (SLA) breaches are both made possible by proactive resource management in cloud services. To that end, several cloud computing providers are turning to Deep Learning (DL) for workload prediction, which can deduce complex multidimensional data. Both the architecture and the quality of the data determine the model's overall quality. That is why it is crucial to use high-quality data while training the model. However, prior research in this area has either relied on an incomplete or biassed data set or has ignored the significance of consistency in conducting fair and reliable analyses. Because of this, DL models' effectiveness is diminished. With the goal of conducting an unbiased analysis, this paper presents a technical analysis of using DL models like RNN, MLP, LSTM, and CNN to leverage the time series characteristics of real-world workloads from the Parallel Workloads A