| Drift Visualization gives better insight into the nature of changes in the data distribution. Cloud trace is dynamic and generated at a very high pace, it’s a serious problem that needs more deep discussions. Drift in cloud resource usage can cause low resource efficiency. In order to achieve optimal resource utilization, cloud provider needs a prediction model based on the data insights. Efficient drift detection optimizes the model prediction. Changes in data can cause these models to reduce their accuracy over a period of time. The focus of this research is to visualize the drift in the cloud at the cluster level. These visualizations will help cloud providers in understanding major factors contributing to the drift. In this paper, Cluster-based visualization using k-means is used to show the drift in the cloud. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.