Power Usage Efficiency (PUE) Optimization with Counterpointing Machine Learning Techniques for Data Center Temperatures

Abstract

The rapid increase in the IT infrastructure has led to demands in more Data Center Space & Power to fulfil the Information and Communication Technology (ICT) services hosting requirements. Due to this, more electrical power is being consumed in Data Centers therefore Data Center power & cooling management has become quite an important and challenging task. Direct impacting aspects affecting the power energy of data centers are power and commensurate cooling losses. It is difficult to optimise the Power Usage Efficiency (PUE) of the Data Center using conventional methods which essentially need knowledge of each Data Center facility and specific equipment and its working. Hence, a novel optimization approach is necessary to optimise the power and cooling in the data center. This research work is performed by varying the temperature in the data center through a machine learning-based linear regression optimization technique. From the research, the ideal temperature is identified with high accuracy based on the prediction technique evolved out of the available data. With the proposed model, the PUE of the data center can be easily analysed and predicted based on temperature changes maintained in the Data Center. As the temperature is raised from 19.73 oC to 21.17 oC, then the cooling load is decreased in the range 607 KW to 414 KW. From the result, maintaining the temperature at the optimum value significantly improves the Data Center PUE and same time saves power within the permissible limits.

CET-4 Listening Test Effect on Listening Learning Based on Machine Learning

“…Machine learning algorithms aim to find patterns autonomously from a class of 3 Wireless Communications and Mobile Computing unknown data and then use this pattern to classify the remaining data or predict the next incoming data in advance. Therefore, the purpose of machine learning is to design algorithms that allow computers to learn autonomously, thereby realizing the application of artificial intelligence [16].…”

Energy Efficient Virtual Machines Placement Over Cloud-Fog Network Architecture

“…In order to combat this, VM consolidation is a technique that involves clustering VMs to servers with computing capacity left, however this decreases the performance of the VM. In this work, VM consolidation is not considered, despite its overall power efficiency improvement [27]. In order to maintain QoS, a VM service maintains a base workload load, this occurs from the CPU utilization used to host the OS, in order to provide the IaaS, SaaS, and PaaS services to users that request it.…”

“…Constraints (27)(28)(29)(30)(31)(32) simulate the placement of a VM in a metro, PON and core networks…”

Data center power management using neural network

“…Customers of cloud services are able to dynamically adjust their resource demands due to the flexibility of cloud resources. However, issues with resource utilization, load imbalance, with excessive power consumption might arise due to changes in resource needs and the pre-defined dimensions of VMs [4] . Hosting Internet-related services with cloud computing is made possible by the physical infrastructure provided by data center networks.…”