Novel Autonomous Software for Enhanced Data Center Operational Efficiency and Botnet Detection

This work presents original research, validated by experimental results on the efficiency and security potential of an optimized and novel approach to an Autonomous Management Distributed System (AMDS) running in a Cloud Computing environment. The AMDS reduces hardware power consumption by autonomously moving virtual servers around a network to balance out hardware loads, as well as being easily configurable and scalable; these are advantages provided through its bespoke software architecture. The novel software module design for the AMDS uses a heuristic detection algorithm. The system also enhances data center security by detecting Botnet activity and preventing the disruption of day-to-day operations. The results demonstrate the AMDS’ potential as an industrial application to be used in modern data centers, both in terms of its ability to reconfigure itself on the fly, resulting in a 14 percent increase in efficiency over its lifetime, and in demonstrating an overall malicious (Botnet) data packet detection rate of over 52 percent (significant for the 5,000 network data samples analyzed by the Botnet software module integrated into the AMDS). Both experiments used for validation were performed in a VMWare run cloud environment; however due to the AMDS’ abstract architecture, this has the potential to interface with any existing cloud management system that exposes an API.