No. 2 (2016)
Published:
2016-06-30
ARTICLES FROM THIS ISSUE
-
Server Workload Model Identification: Monitoring and Control Tools for Linux
Abstract
Server power control in data centers is a coordinated process carefully designed to reach multiple data center management objectives. The main objectives include avoiding power capacity overloads and system overheating, as well as fulfilling service-level agreements (SLAs). In addition to the primary goals, server control process aims to maximize various energy efficiency metrics subject to reliability constraints. Monitoring of data center performance is fundamental for its efficient management. In order to keep track of how well the computing tasks are processed, cluster control systems need to collect accurate measurements of activities of cluster components. This paper presents a brief overview of performance and power consumption monitoring tools available in the Linux systems.
-
Energy-saving Algorithms for the Control of Backbone Networks: A Survey
Abstract
The rapid growth of energy demand by wired IP networks can be mitigated on hardware and software levels. While upgrading to more efficient transmission media still brings biggest savings, we take a look here at power-saving algorithms that combine the capability of setting networking equipment in arbitrary energy states which, combined with profound knowledge of the network traffic matrix, leads to considerable complex optimization problem formulations. Alternatively, lightweighted heuristic approaches are presented, built on much simpler network model but still capable to perform energy-efficient traffic engineering.
-
New Developments in a Two-criteria Approach to Dynamic Power Management in Energy-aware Computer Networks
Abstract
In the paper authors continue the development of a model of dynamic power management in energy-aware computer networks, where two criteria: energy consumption and the quality of service are considered. This approach is appropriate when the routing problem with fixed demands is inadmissible. The formulation introducing edge indices is modified and tests on problems of different sizes are performed.
-
Preconditioned Conjugate Gradient Method for Solution of Large Finite Element Problems on CPU and GPU
Abstract
In this article the preconditioned conjugate gradient (PCG) method, realized on GPU and intended to solution of large finite element problems of structural mechanics, is considered. The mathematical formulation of problem results in solution of linear equation sets with sparse symmetrical positive definite matrices. The authors use incomplete Cholesky factorization by value approach, based on technique of sparse matrices, for creation of efficient preconditioning, which ensures a stable convergence for weakly conditioned problems mentioned above. The research focuses on realization of PCG solver on GPU with using of CUBLAS and CUSPARSE libraries. Taking into account a restricted amount of GPU core memory, the efficiency and reliability of GPU PCG solver are checked and these factors are compared with data obtained with using of CPU version of this solver, working on large amount of RAM. The real-life large problems, taken from SCAD Soft collection, are considered for such a comparison.
-
A Novel GPU-Enabled Simulator for Large Scale Spiking Neural Networks
Abstract
The understanding of the structural and dynamic complexity of neural networks is greatly facilitated by computer simulations. An ongoing challenge for simulating realistic models is, however, computational speed. In this paper a framework for modeling and parallel simulation of biological-inspired large scale spiking neural networks on high-performance graphics processors is described. This tool is implemented in the OpenCL programming technology. It enables simulation study with three models: Integrate-andfire, Hodgkin-Huxley and Izhikevich neuron model. The results of extensive simulations are provided to illustrate the operation and performance of the presented software framework. The particular attention is focused on the computational speed-up factor.
-
Forecasting Stock Price using Wavelet Neural Network Optimized by Directed Artificial Bee Colony Algorithm
Abstract
Stock prediction with data mining techniques is one of the most important issues in finance. This field has attracted great scientific interest and has become a crucial research area to provide a more precise prediction process. This study proposes an integrated approach where Haar wavelet transform and Artificial Neural Network optimized by Directed Artificial Bee Colony algorithm are combined for the stock price prediction. The proposed approach was tested on the historical price data collected from Yahoo Finance with different companies. Furthermore, the prediction result was found satisfactorily enough as a guide for traders and investors in making qualitative decisions.
-
A Cloud-aided Group RSA Scheme in Java 8 Environment and OpenStack Software
Abstract
In this paper a RSA based security system enabling the group of users to upload the single masked message to the cloud environment is proposed. Data stored are encrypted using RSA algorithm. The data receiver is able to encrypt the message retrieved from the cloud environment using private key. Two different separate RSA systems are used. The presented approach is divided into three parts. First, an RSA key is generated for each sender. Then masking the message by newly chosen mask proposed individually by every member, additionally encrypted by individual RSA private key of each member is proceed. Next, encrypting the gathered message inside the cloud environment, using the public key of the receiver is executed. In the third step, the message is decrypted by the receiver using his private RSA key. The scheme reduces the computational load on users side and transfers calculations and storage effort to the cloud environment. The proposed algorithm was developed for storing and sending the data that originally are produced by a group of users, but the receiver of the data is single. It was implemented using Java 8 and OpenStack software. Numerical test of different key length for RSA are presented.
-
Graph-based Forensic Analysis of Web Honeypot
Abstract
Honeypot still plays an important role in network security, especially in analyzing attack type and defining attacker patterns. Previous research has mainly focused on detecting attack pattern while categorization of type has not yet been-comprehensively discussed. Nowadays, the web application is the most common and popular way for users to gather information, but it also invites attackers to assault the system. Therefore, deployment of a web honeypot is important and its forensic analysis is urgently required. In this paper, authors propose attack type analysis from web honeypot log for forensic purposes. Every log is represented as a vertex in a graph. Then a custom agglomerative clustering to categorize attack type based on PHP-IDS rules is deployed. A visualization of large graphs is also provided since the actual logs contain tens of thousands of rows of records. The experimental results show that the proposed model can help forensic investigators examine a web honeypot log more precisely.
-
Network Function Virtualization: Mitigating the Impact of VoLTE on the Policy and the Charging System
Abstract
Needless to say that telecommunications’ operators are showing increasing interests toward solving the dilemma of optimizing network resources while offering state-of-the-art quality of service. Recently, operators have shown an increasing interest to investigate solutions for better control on network utilization, service usage and monetization. They also noticed a significant growth in Diameter signaling and more specifically in signaling related to policy management. A massive introduction of Voice over LTE (VoLTE) service will have a significant impact on the systems handling policy signaling, as VoLTE will reshape the landscape of Long Term Evolution (LTE) policies and completely change the way policy management works. However, this massive approach is meant to provide significant competitive advantages for operators offering LTE services and still require circuit-switched network to provide voice service. The biggest challenge for those operators is to find an appropriate solution, scalable enough to handle the unpredictable growth of Diameter signaling. In this paper, a model, based on Network Function Virtualization (NFV) technology is proposed, able to address the challenges of massively introducing VoLTE, without impacting existing services and without jeopardizing current revenues. In presented approach, the standard VoLTE call flows, referenced user’s behavior and latest experiments’ results on NFV technology are used.
-
Application of Social Network Inferred Data to Churn Modeling in Telecoms
Abstract
The subject of this work is the use of social network analysis to increase the effectiveness of methods used to predict churn of telephony network subscribers. The social network is created on the basis of operational data (CDR records). The result of the analysis is customer segmentation and additional predictor variables. Proposed hybrid predictor employs set of regression models tuned to specific customer segments. The verification was performed on data obtained from one of the Polish operators.
-
Similarity Index based Link Prediction Algorithms in Social Networks: A Survey
Abstract
Social networking sites have gained much popularity in the recent years. With millions of people connected virtually generate loads of data to be analyzed to infer meaningful associations among links. Link prediction algorithm is one such problem, wherein existing nodes, links and their attributes are analyzed to predict the possibility of potential links, which are likely to happen over a period of time. In this survey, the local structure based link prediction algorithms existing in literature with their features and also the possibility of future research directions is reported and discussed. This survey serves as a starting point for beginners interested in understanding link prediction or similarity index algorithms in general and local structure based link prediction algorithms in particular.
-
Antenna Arrays Focused on Broadband Signals
Abstract
Broadband and ultra-wideband signals are increasingly used in modern radio systems. Traditional performance of evaluation antennas operating with narrowband signals are not always adequately reflect the characteristics of broadband antennas, at least in view of the frequency dependence of the antenna pattern. Accounting for broadband signals the antennas becomes important in the low-frequency range of the spectrum. Systems using these types of signals may include control of the atmosphere and measuring its frequency-selective properties in the range meter and decameter wavelengths. Possibility of spatial selection based on focusing of broadband signals in this case promises to implement a number of additional features. Therefore, it is important to evaluate the properties of antennas based on the spectral content of the signal, as well as taking into account the ways of its processing in the receiving equipment. Consideration features of functioning the antenna array, focused on broadband signal is devoted to this article.
-
Radio Photonic Systems for Measurement of Instantaneous Radio Frequency with Amplitude-phase Modulation of Optical Carrier
Abstract
In this article questions related to the development instantaneous radio frequency measurement system based on application in them original ways of the amplitude-phase modulation transformation of single-frequency optical carrier by a radio signal in symmetric two-frequency and measuring “frequency-amplitude” transformation in fiber Bragg grating with special profile are considered. Such systems have broad prospects for use in telecommunications, military systems and for environmental monitoring.
-
Support Vector Machine based Decoding Algorithm for BCH Codes
Abstract
Modern communication systems require robust, adaptable and high performance decoders for efficient data transmission. Support Vector Machine (SVM) is a margin based classification and regression technique. In this paper, decoding of Bose Chaudhuri Hocquenghem codes has been approached as a multi-class classification problem using SVM. In conventional decoding algorithms, the procedure for decoding is usually fixed irrespective of the SNR environment in which the transmission takes place, but SVM being a machinelearning algorithm is adaptable to the communication environment. Since the construction of SVM decoder model uses the training data set, application specific decoders can be designed by choosing the training size efficiently. With the soft margin width in SVM being controlled by an equation, which has been formulated as a quadratic programming problem, there are no local minima issues in SVM and is robust to outliers.
-
An Efficient Early Iteration Termination for Turbo Decoder
Abstract
Turbo code finds wide applications in mobile communication, deep space communication, satellite communication and short-range communication despite its high computational complexity and iterative nature. Realizing capacity approaching turbo code is a great achievement in the field of communication systems due to its efficient error correction capability. The high computational complexity associated with the iterative process of decoding turbo code consumes large power, introducing decoding delay, and reducing the throughput. Hence, efficient iteration control techniques are required to make the turbo code more power efficient. In this paper, a simple and efficient early iteration termination technique is introduced based on absolute value of the mean of extrinsic information at the component decoders of turbo code. The simulation results presented clearly show that the proposed method is capable of reducing the average number of iterations while maintaining performance close to that of fixed iteration termination. The significant reduction in iteration achieved by the method reduces decoding delay and complexity while maintaining Bit Error Rate performance close to standard fixed iteration turbo decoder.
