Proactive Data Centre & Network Room Overheating Management System (DCNROMS) Utilizing Open Source Software and Tools
Vikas Maurya, Sandeep Yadav, K. K. Bachhil, Harish Chouhan, S. Chaudhari, S. S. Tomar, Alpana Rajan, Anil Rawat · 2019
Data centre equipment are very sensitive and susceptible to environmental damage from excessive heat. Power outages can lead to overheated servers in a matter of minutes. It is therefore mandatory to monitor the ambient temperature of the data centre or network room continuously and take necessary action if temperature exceeds the threshold limit. Manual 24x7 monitoring requires additional man power and is not an efficient approach. The paper presents an effective and cost efficient mechanism to automate this process using open source software's and tools. We have utilized the inbuilt temperature sensing capabilities of the managed network switches to build such a system. Thus the system has been designed without the use of additional temperature monitoring sensors. The system has been implemented in our organization and has helped in reducing the failure rate of the network switches from 2% to 0%, thus helping in maintaining 99.98% uptime of the network.