سال انتشار: ۱۳۹۱

محل انتشار: کنفرانس بین المللی مدل سازی غیر خطی و بهینه سازی

تعداد صفحات: ۷

نویسنده(ها):

Mohammad Abedi – Department of Electrical Engineering and Computer Science at Iran University of Science and Technology
Mehdi Darbandi –

چکیده:

Cloud is a virtual image about some amount of undefined powers, that is widespread and had unknown power and inexact amount of hardware and softwareconfigurations, and because of we haven’t any information about clouds location and time dimensions and also the amounts of its sources we tell that Cloud Computing [1]. Thistechnology presents lots of abilities and opportunities such as processing power, storage and accessing it from everywhere,supporting, working – team group – with the latest versions of software and etc., by the means of internet. Cloud computing isreliable services that presenting by next generation data centers and from internet, they based on virtual technologies and computing methods [4]. In recent decades the internet has very deep influences in human lives and also in offer and demand markets; lots of people for trading, reading the news,seeing the movies or to play online games go through theinternet. As our daily needs from the internet became more, sothat the processing power and amount of storage data and filesshould increase significantly [8]. With remarkable developments of communication and information technology inrecent days, we consider processing and processing power as the fifth vital component in human lives (after water,electricity, gas and telephone) [9]. These days’ lots oftechnologies migrate from traditional systems into cloud and similar technologies; also we should note that cloud can beused for military and civilian purposes [3]. On the other hand, in such a large scale networks we should consider thereliability and powerfulness of such networks in facing withevents such as high amount of users that may login to their profiles simultaneously, or for example if we have the ability topredict about what times that we would have the most crowd in network, or even users prefer to use which part of the CloudComputing more than other parts – which software or hardware configuration. With knowing such information, we can avoid accidental crashing or hanging of the network that may be cause by logging of too much users. In this paper we propose Kalman Filter that can be used for estimating theamounts of users and software’s that run on cloud computing or other similar platforms at a certain time. After introducing this filter, at the end of paper, we talk about some potentials of this filter in cloud computing platform. In this paper we demonstrate about how we can use Kalman filter in estimatingand predicting of our target, by the means of several examples on Kalman filter. Also at the end of our paper we demonstrate the sensitivity of Kalman filter on cloud computing platform, and with an example we show the effectiveness of this filter.