An Improvement in Power Management in green Computing using Neural Networks

The green computing is the technology which is based on the environmental use of computer related resources. The computer related resources includes processing units, storage units etc. In such type of technology the energy consumption is the main concern. Green computing wills leads to reduce in resource consumption and electronic waste. The new technology of green computing will also include cloud computing services, grid computing services. In the past decades, many techniques have been proposed for the energy conservation in green computing. In this paper, we are proposing new technique for energy conservation in green computing. This novel technique is based on neural networks. The neural network is having capability of learning from the past experiences. The dynamic clustering approach is used with the neural networks for the energy conservation. The proposed technique is implemented in NS2 and simulation results are shown in the graphical form. The green computing is basically a technology which is based on the environmental responsible way to reduce power and environmental waste. In the present time, we are making efforts to make everything greener. This technology is related to reduce or managing impact of IT systems which basically includes materials and resources required for equipment, energy and also materials used in OS, potential health effects on humans from using equipment, and responsibility for the waste products that are created from IT systems. The green technology mainly focuses on super computers and cluster system which harm the green environment badly. The green computing is similar to green chemistry. In the approach of green computing we mainly focus on to design such systems which are having minimum effect to environment. The products supports reduce the use of hazardous materials, maximum energy efficient during the products lifetime and promote the recyclability of defunct products and factory waste. Artificial neural network is composed of interconnecting of artificial neurons. Artificial neural networks may either be used to gain understanding of biological neuron or to solve artificially intelligence problems without creating any model (8). Biological neural networks are made up of real biological neurons that are connected or functionally related in nervous system (wiki).neural network has been motivated from human brain. The brain is highly complex, parallel computer and nonlinear. It has capability to organize its constituent's structure known as neurons to perform complex computations. It is faster than digital computer exit in today's world (8). It is an adaptive in nature that changes its structure based on internal and external information that flows through the networks. Dynamic clustering is an energy efficient algorithm. Energy dissipation of the network can be reduced by using clustering algorithms. The energy consumption of nodes is depends upon the transmission distance, optimal routing protocols and amount of data to be transmitted. In cluster based networks, cluster heads (CH) meets these requirements 1) same adjacent sensors are grouped into a cluster. 2) High energy resources 3) Network should be distributed.