Norliza Zaini

Virtual machine migration implementation in load balancing for Cloud computing

The demand for Cloud computing usage is increasing day by day due to the advantages and facilities it offers. The fast growing rate of the usage of large-scale computing machines on cloud platform has resulted in increased consumption of energy and emission of carbon. Such negative effects should be curbed for a more environmental-friendly computing platform, e.g. a green cloud computing platform. Scheduling is one of the aspects that can be looked into in regards to the improvement of the whole cloud architecture, specifically in the scope of resource management. One of the important features that need to be optimized in regards to scheduling is the load balancing process that emphasizes on optimal resource utilization, maximum throughput, maximum response time and prevention of overload. One way to optimize the load balancing process is by implementing the migration of virtual machines across multiple hosts, in which utilization of CPU resources can be optimized. The appropriate use of two different classes of resources e.g. low-powered and high-powered machines based on their Million Instructions Per Second (MIPS) metrics, will lead to minimal process execution time if both types of resources are being efficiently used and effectively mapped to suitable types of processing. According to different machine performance, the migration of virtual machines will be computed based on the current CPU utilization following three different conditions, i.e. when the CPU usage reaches its 90%, 10% and 0% marker. Based on this idea, a general algorithm in activating the virtual machine migration is proposed, in which the overall load balancing process could be improved.

Remote monitoring system based on a Wi-Fi controlled car using Raspberry Pi

Developing surveillance and monitoring systems can be quite challenging at times, since the systems should be designed with consideration of the environment to be monitored. Good surveillance systems need to have dynamic features, e.g. monitoring cameras that are mobile and can move around the area being monitored. One scenario that requires such mobile monitoring is the surveillance of a large building with many levels, which would incur a high cost to install many cameras at many places. Monitoring such a large area would also be a challenge for the security officers, as they will need to spend too much time to patrol covering all places. Other scenarios that require dynamic surveillance systems include dangerous areas, e.g. areas with explosion and fire risk, or those contaminated with toxic gases. Another use case includes areas that cannot be accessed by humans. To address these challenges, we propose a dynamic monitoring system based on a moving camera carried by a Wi-Fi remote controlled car. The remote-controlled (RC) car is able to move on its own in an area remote from the operator. Users can be in the next room, the next building, or even in another country while controlling the movement of the car via the Internet. Some sensing mechanism is needed to help the user to locate the current location of the car and effectively navigate. Thus, the camera that is attached to the car also act like the eyes showing the frontal view of the car to be seen by the user via a browser in real-time. This system is using Wi-Fi as the only communication medium to connect the car to the server. The web application which provides the interface for the user to interact with the car also provides live streaming views in the vicinity to be monitored, i.e. via some other CCTV cameras installed in the area.

Support vector machine for classification of stress subjects using EEG signals

Stress is a mental condition that can effects the brain electrical activity to be different from the normal state. This brain cognitive change can be measured using EEG. The objective of this paper is to classify stress subjects based on EEG signal using SVM. The data which are used to represent stress subjects were taken from the residents of Pusat Darul Wardah; a shelter centre for troubled women. SVM is used to classify the EEG Alpha band data for Power Spectral Density and Energy Spectral Density. Using 5-fold cross validation, the classification rate are 83.33% for ESD data using RBF kernel function.

An online system to support collaborative knowledge acquisition for ontology development

Ontology development is an important practice required in any knowledge engineering process. Before knowledge can be integrated and put into good use by computer applications, knowledge acquisition and modeling has to be done. This signifies the adoption of ontology; which is defined as conceptualization or formal specifications of a domain; as the core framework for a knowledge-based system. For faster ontology development process, it is normally done in a collaborative manner and the challenge of having an environment that supports collaborative work is manifold. This includes the challenge of forming a collaborative procedure in acquiring and modeling the knowledge, allocating a platform that is accessible to geographical distributed parties and to handle update conflicts during the development. This paper presents a system in supporting initial collaborative ontology development works between ontology developers without requiring the presence of domain experts, aiming at accelerating the ontology development process. This system is designed in such a way that it incorporates and formalizes a novel collaborative knowledge acquisition and modeling procedure.

Android-based application for real time energy monitoring of domestic electricity

The electricity demand keeps on increasing year by year. In Malaysia, energy consumption has increased rapidly from an average of 5% in 1980s and 12% in 2009. Due to this increasing trend, government has encouraged users to reduce their electricity consumption. One of the way is by introducing the smart meter as a solution, where the smart meter device is used to display the total of energy consumption by users. In this context, we have developed an online-based energy smart monitoring system for domestic electricity consumption. This system offers a beneficial feature, where the user is able to monitor which part of a house consumes more power energy than others. In recording the electricity usage, smart meter microcontroller is installed to continually measure data for different areas of a house. The captured electricity usage data is uploaded by the microcontroller to an online database, which can then be viewed by users on their Android devices. Through this monitoring system, the objectives are achieved such that the users are able to know how much energy are being consumed for a particular area in the house. The Android application provides an interface to show the analysis of electricity consumption for different areas of a house in terms of voltage, current, power consumption and the corresponding pricing for a specified duration.

User profile based product recommendation on android platform

When visiting malls or stores, users are usually served with a variety of products from different brands. Some products clearly reflect their suitability to the user, while a few others do not portray their suitability to the user. For instance, some foods are not suitable for Muslims, i.e. when the food contains ingredients that are not halal. Such case is particularly troublesome in the event where there are no labels or sign on the food products indicating whether they are halal or vegetarian. In such a case, the user may need to scrutinize the foods ingredients one by one. One of the ways to overcome these problems is by searching the facts on this particular matter on the web. However, this solution consumes the users time especially to search for the products details. To resolve this issue, in this paper we present an Android-based application called the User Profile Based Product Recommender. Basically, this application provides recommendations to its user based on the users profile, e.g. gender, religion and medical condition. According the the recommendations being generated, the application will be able to match suitable products to the user. Additionally, the application provides an easy means in identifying the products, i.e. by scanning the products barcode. By having such a simple-to-use application, this application will be a helpful tool in facilitating the users to be wiser in choosing the right products. Android platform is chosen as the hosting platform due to its nature, which is very close and personal to the user.

Dynamic encoding of binaural beats for brainwave entrainment

This paper shows the development of the dynamic binaural beat encoding using JAVA program. Binaural beat is an auditory phenomenon that has been used worldwide to alter our brain state, to get to the desired state, e.g. Relax. However, most of the Binaural Beat system designed is a static binaural beat on which the user is not able to alter the frequency of the audio. The binaural beat frequency listened by the user might not be suitable to the users current brain state. Such incompatibility may cause the user to feel uncomfortable and may also cause dizziness while listening to the audio being played. Hence, dynamic encoding of binaural beats is designed and implemented to replace the existing static encoding of binaural beats. The GUI of the system has two options, Manual Configuration and Automatic Configuration on which the user able to choose the base frequency and the offset frequency as desired. The developed system has been tested and validated in the lab to show its ability to generate the binaural beat frequency.

Web-based adaptive audio-therapy recommender system

This paper presents an online system that aims to recommend suitable audio-therapy to improve mental states of a user. It is designed and implemented with a capacity to analyse users current mental states and based on the acquired result, recommend a suitable audio-therapy for the user. The web based system was developed with a number of components. The core engine of the system is an online questionnaire component used to determine the users current mental states and an audio-therapy matching component. The latter is driven by an algorithm created to match the users score derived from the answered questionnaire to suitable therapy audios, recommended for the purpose of improving the users mental states. The system was implemented using database-driven PHP server script, enabling dynamic functionalities and features offered to both end-users and psychiatrists.

Semantic-based Bayesian Network to determine correlation between binaural-beats features and entrainment effects

In coping with hectic everyday lives, people have tried many different ways to reduce stress and depression. Such effort has also lead to meditation practice, which is believed and proven able to help. While meditation is beneficial, for some people, meditation is just hard to perform. These people may switch to other alternatives that give the same effects as meditation, such as through binaural beats entrainment. Many studies on brainwave entrainment have demonstrated that the brain responds by synchronizing its own electrical cycles to the same rhythm of the stimulating binaural-beats audio. The results of binaural beats entrainment towards the brainwave can be determined by monitoring the EEG readings, which can be analyzed to capture the altered brainwave patterns and qualities they exhibit. In relation to the monitoring process, our work focuses on capturing and analyzing the correlations between different binaural beats features to resulting EEG and perceived mental states. A general methodology is presented while detailing further on the proposed Semantic-based Bayesian Network Engine, which is the core mechanism employed in capturing the correlations. This novel approach is proposed firstly due to the well-known capability of Bayesian Network in modeling the elements of causal and effects. Secondly, with the introduction of semantic notion, the engine is enhanced even more for allowing dynamic-construction of Bayesian Network based on its semantics.

Throughput analysis of LAN and WAN network based on socket buffer length using JPerf

Cluster computing involves connection between a server and multiple hosts in one single environment and multi-cluster in inter-cluster environments. One of the crucial factors of a clustering is its network communication. Relying for support of a server-client internetworking, cluster computing uses the TCP/IP socket communication mechanism extensively. Mechanism that controls the data transfer interruptions between a server and client is referred to as a flow control. Theoretically, if the receive window size for TCP/IP buffers is too small, the receive window buffer is frequently overrun, and the flow control mechanism stops the data transfer until the receive buffer is empty. Thus, as a result flow control activity contributes to unnecessary additional network latency which is unacceptable in any server-client networking especially in clustering environment where it involves highly reliable data transfer. Among the solutions to this problem is by adjusting buffer size to avoid or reduce the potential for flow control to occur. This paper describes the analysis of TCP/IP socket buffer length in Local Area Network (LAN) and Wide Area Network (WAN) clustering. It details the process and flow of setting up the environment, implementing as well as analyzing the throughput performance of the measured buffering.