A novel autonomous management distributed system for cloud computing environments

Abstract

This paper describes a novel modular design of an autonomous management distributed system (AMDS) for cloud computing environments and it presents its implementation with the Scala programming language. The AMDS was designed from the ground up with distributed deployment, modularity and security in mind, using a full object oriented approach. A key feature of this system is the ability to gather and store information from various networking and monitoring devices from within the same computing cluster. Another key feature is the ability to intelligently control VMWare vSphere local instances based on analysis of collected data and predefined parameters. vSphere in turn, once it receives commands from the AMDS, proceeds to issue instructions to multiple locally monitored ESXi severs in order to maximize energy efficiency, reduce the carbon footprint and minimize running costs. The predefined parameters are based on results from a previous paper written by the authors. The AMDS has been deployed on the authors' test bed and is currently running successfully. Test results show highly potential industrial applications in datacenter energy management and lowering of operating costs.

A software approach to improving cloud computing datacenter energy efficiency and enhancing security through Botnet detection

“…The main driver of this increase over time is the system's ability to reconfigure itself on the fly based on changes happening within and around the virtual environment it resides in, novel design features presented in [4]. These results demonstrate what can be achieved using a purely software approach to an existing datacenter hardware infrastructure.…”

“…The novelty consists of the software's ability to reconfigure itself on the fly based on live network readings [4]. The authors also critically developed a novel method to prevent, detect, and stop network intrusions and malicious behavior in a cloud environment [5].…”

Analytical Modeling of Reactive Autonomic Management Techniques in IaaS Clouds

“…Reactive autonomic management techniques are recently gaining attention as a way to deal with this kind of situations in a dynamic way [7], [8]. Through this kind of techniques, IaaS cloud managers can configure the system to detect VM request arrival bursts and react autonomously by changing the system configuration.…”