Journal
JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING
Volume 152, Issue -, Pages 88-97Publisher
ACADEMIC PRESS INC ELSEVIER SCIENCE
DOI: 10.1016/j.jpdc.2021.02.018
Keywords
Cloud computing; Cloud-based service; Predicting resource consumption; Adaptive planning; Machine learning
Categories
Funding
- Polish Ministry of Science and Higher Education, Poland
Ask authors/readers for more resources
The article discusses the issue of resource planning when using cloud computing resources for services, with a focus on addressing overestimation of resource utilization. Implementing machine learning methods for cloud resource reservation planning has shown promising results for both short- and long-term reservations.
The problem of using cloud computing resources for services is related to planning the amount of resources needed and their subsequent reservation. This problem occurs both on the side of the customer who tries to minimize the cost of the service and on the side of the cloud provider who wants to make the best use of existing infrastructure without introducing any modifications. In our article, we want to show how the problem of overestimating the utilization of resources for services which use cloud computing can be handled. Solving this problem will allow significant savings to be made by both the customer and the cloud infrastructure provider. The system we have developed demonstrates the considerable utility of machine learning methods when planning cloud resource reservation for network services. The models proposed, which use a multilayer perceptron, have yielded good results for both short- and long-term reservations. (C) 2021 Elsevier Inc. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available