Intelligent Admission Control Manager using Decision Tree Classifier in Fog Integrated Cloud
6 Pages Posted: 30 Jan 2020
Date Written: January 29, 2020
Abstract
Internet of Things (IoT) and related devices are producing the data, which are very large in volume, variety, and velocity. There are many time critical applications, and by the time the data is sent to cloud for analysis/execution, the opportunity to act might be gone. Consequently, we may face problem either financially or other way round where cost is intangible. So time sensitivity matters a lot, and such criticality could be achieved through smart admission control manager at the nearest node usually in form of fog layer architecture. In this paper, a smart admission control manager is being proposed which has the primary responsibility of placing the jobs based on the request parameters such as CPU, Memory, Disk Space etc. besides some other categorical parameters e.g. priority of request, time-sensitivity of request etc. Post-processing, data may be forwarded for further analysis to the Cloud. The proposed work applies a machine learning technique which first classifies the jobs into two or three categories, based on the requirement of the domain. Further, it will decide on job placement to the respective nodes using decision tree classifier.
Keywords: Fog Computing, Cloud Computing, Internet of Things (IoT), Machine Intelligence Resource Provisioning
Suggested Citation: Suggested Citation