Skip to main content
main-content

Über dieses Buch

Based on a rigorous selection of submissions to The 29th International Symposium on Computer and Information Sciences (ISCIS 2014), this books includes some of the most recent ideas and technical results in computer systems, computer science, and computer-communication networks. It offers the reader a timely access to innovative research and advances in computing and communications from many different areas of the world.

The topics covered include (but are not limited to) computer architectures and digital systems, algorithms, theory, software engineering, data engineering, computational intelligence, system security, computer systems and networks, performance modeling and analysis, distributed and parallel systems, bioinformatics, computer vision and significant applications such as medical informatics and imaging.

The 29th International Symposium on Computer and Information Sciences (ISCIS 2014) took place in Krakow Old City, Poland on October, 27–8, 2014.

Inhaltsverzeichnis

Frontmatter

Erratum to: Information Sciences and Systems 2014

Tadeusz Czachórski, Erol Gelenbe, Ricardo Lent

Wireless and Cognitive Networks

Frontmatter

New Channel Access Approach for the IEEE 802.15.4 Devices in 2.4 GHz ISM Band

The number of indoor wireless communication devices and technologies employing the ISM band is increasing day by day. As a consequence, co-existence is a big challenge for tiny, IEEE 802.15.4-based and resource-constrained devices. In this paper, 2.4 GHz ISM band aggregated traffic is analyzed and a novel channel access approach is proposed based on the cognitive radio concepts in order to increase free channel access performance. The performance is evaluated by using the real-world RF signal strength measurements of an indoor IEEE 802.15.4 node with the presence of many IEEE 802.11 interferers. The performance evaluation gives promising results considering free channel access performance.

Tolga Coplu, Sema F. Oktug

A Parametric Study of CPN’s Convergence Process

The Cognitive Packet Network routing algorithm is a routing algorithm for “self-aware” networks, which continuously monitors the state of the network and is able to respond to changes in network conditions with low latency. In particular, the monitoring and exploration process can be guided by Random Neural Networks to provide the best performance for the lowest search overhead. CPN and RNN have been the focus of several research papers, however these provide little to no detail on how parameters are set. This paper attempts to bridge this gap in the literature by proposing a bench-test experiment of CPN’s initial knowledge gathering process (convergence), whilst modifying the values assigned to key parameters. We discover that one of the parameters controls CPN’s tendency to either produce low-quality results very quickly, but with little improvement over time; or a “slow-but-steadily improving” solution. We also find that another parameter can save some search overhead with minimal impact on the resulting paths’ quality.

Antoine Desmet, Erol Gelenbe

Multi-cell Resource Block Allocation Framework

We propose to combine the beamforming technique with Resource Block (RB) allocation algorithms to improve the performance in OFDMA networks. With MIMO antennas, the beamforming technique improves the received signal power which increases the RB’s capacity and reduces the neighbouring cell users’ interference. When the inter-cell interference channels are known, the beamforming parameters could be applied to the iterative scheduling methods to enhance the performance of the beamforming technique, hence increasing the total system throughput.

Fan Huang, Véronique Veque, Joanna Tomasik

An Implementation of Voice Over IP in the Cognitive Packet Network

Voice over IP (VOIP) has strict Quality of Service (QoS) constraints and requires real-time packet delivery, which poses a major challenge to IP networks. The Cognitive Packet Network (CPN) has been designed as a QoS-driven protocol that addresses user-oriented QoS demands by adaptively routing packets based on online sensing and measurement. This paper presents our design, implementation, and evaluation of a ‘Voice over CPN’ system where an extension of the CPN routing algorithm has been developed to support the needs of voice packet delivery in the presence of other background traffic flows with the same or different QoS requirements.

Lan Wang, Erol Gelenbe

A Cooperative Emergency Navigation Framework Using Mobile Cloud Computing

The use of wireless sensor networks (WSNs) for emergency navigation systems suffer disadvantages such as limited computing capacity, restricted battery power and high likelihood of malfunction due to the harsh physical environment. By making use of the powerful sensing ability of smart phones, this paper presents a cloud-enabled emergency navigation framework to guide evacuees in a coordinated manner and improve the reliability and resilience in both communication and localization. By using social potential fields (SPF), evacuees form clusters during an evacuation process and are directed to egresses with the aid of a Cognitive Packet Networks (CPN)-based algorithm. Rather than just rely on the conventional telecommunications infrastructures, we suggest an Ad hoc Cognitive Packet Network (AHCPN)-based protocol to prolong the life time of smart phones, that adaptively searches optimal communication routes between portable devices and the egress node that provides access to a cloud server with respect to the remaining battery power of smart phones and the time latency.

Huibo Bi, Erol Gelenbe

Data and Image Analytics

Frontmatter

Integer Linear Programming Solution for the Multiple Query Optimization Problem

Multiple Query Optimization (MQO) is a technique for processing a batch of queries in such a way that shared tasks in these queries are executed only once, resulting in significant savings in the total evaluation. The first phase of MQO requires producing alternative query execution plans so that the shared tasks between queries are identified and maximized. The second phase of MQO is an optimization problem where the goal is selecting exactly one of the alternative plans for each query to minimize the total execution cost of all queries. A-star, branch-and-bound, dynamic programming (DP), and genetic algorithm (GA) solutions for MQO have been given in the literature. However, the performance of optimal algorithms, A-star and DP, is not sufficient for solving large MQO problems involving large number of queries. In this study, we propose an Integer Linear Programming (ILP) formulation to solve the MQO problem exactly for a large number of queries and evaluate its performance. Our results show that ILP outperforms the existing A-star algorithm.

Tansel Dokeroglu, Murat Ali Bayır, Ahmet Cosar

A Graphical Model Approach for Multi-Label Classification

Multi-Label (ML) classification problem is the assignment of many labels to a given sample from a fixed label set. It is considered as the more general version of the Multi-Class (MC) classification problem and its practical application areas vary from medical diagnosis to paper keyword selection. The general structure of an ML classification system involves transforming the problem into simpler MC and Single-Class (SC) problems. One such method is the Binary Relevance (BR) method that treats each label assignment as an independent SC problem, which makes BR systems scalable, but not accurate for some cases. This paper addresses the label independence problem of BR by assuming the outputs of each SC classifiers as observation nodes of a graphical model. The final label assignments are obtained by standard powerful Bayesian inference from the unobservable node. The proposed system was tested on standard ML classification datasets that produced encouraging results.

Meltem Cetiner, Yusuf Sinan Akgul

Ground Plane Detection Using an RGB-D Sensor

Ground plane detection is essential for successful navigation of vision based mobile robots. We introduce a very simple but robust ground plane detection method based on depth information obtained using an RGB-Depth sensor. We present two different variations of the method: the simplest one is robust in setups where the sensor pitch angle is fixed and has no roll, whereas the second one can handle changes in pitch and roll angles. Our comparisons show that our approach performs better than the vertical disparity approach. It produces accurate ground plane-obstacle segmentation for difficult scenes, which include many obstacles, different floor surfaces, stairs, and narrow corridors.

Doğan Kırcalı, F. Boray Tek

An Iterated Local Search Platform for Transportation Logistics

Recent technological advances in optimization and transportation have enabled the development of efficient tools that support the decision-making process of logistic managers. The aim of this paper is to present a decision support system (DSS) that integrates an Iterated Local Search (ILS) for solving the Vehicle Routing Problem (VRP). Its clear design allows an efficient exploration of the solution by the decision-maker. The computational experiments show that the ILS is very competitive in comparison to state-of-the-art algorithms.

Takwa Tlili, Saoussen Krichen

On Fuzzy Extensions to Energy Ontologies for Text Processing Applications

Ubiquitous application areas of domain ontologies include text processing applications like categorizing related documents of the domain, extraction of information from these documents, and semantic search. In this paper, we focus on the utilization of two energy ontologies, one for electrical power quality and the second for wind energy, within such applications. For this purpose, we present fuzzy extensions to these domain ontologies as fuzziness is an essential feature of the ultimate forms of the ontologies to enhance such text analysis applications in the energy domain. A text categorization system for scholarly articles based on the extended power quality ontology is also presented for illustrative purposes together with its performance evaluation results.

Dilek Küçük, Doğan Küçük, Adnan Yazıcı

Evaluating Quality of Dispersion Based Fixation Detection Algorithm

Information hidden in the eye movement signal can be a valuable source of knowledge about a human mind. This information is commonly used in multiple fields of interests like psychology, medicine, business, advertising, or even software developing. The proper analysis of the eye movement signal requires its elements to be extracted. The most important ones are fixations—moments when eyes are almost stable and the brain is acquiring information about the scene. There were several algorithms, aiming at detecting fixations, developed. The studies presented in this paper focused one of the most common dispersion-based algorithms—I-DT one. The various ways of evaluating its results were analyzed and compared. Some extensions in this algorithm were made as well.

Katarzyna Hareżlak, Paweł Kasprowski

An Evaluation of Iris Detection Methods for Real-Time Video Processing with Low-Cost Equipment

The purpose of this work is to accomplish a study aiming the construction of an eye tracking and iris detection system, based on images obtained from a low-cost webcam. The main objective of the paper is to conduct a comparison between three computer vision approaches for iris detection, trying to identify the more suitable method for application in the aforementioned low-cost eye tracking system. The methods which have achieved the best detection rates were the Projection and Thresholding, however, all of them offer possibilities for application in real-time processing and improvement.

Andrey Kuehlkamp, Cristiano Roberto Franco, Eros Comunello

Traffic Measurement and Analysis

Frontmatter

Principles of Pervasive Cloud Monitoring

Accurate and fine-grained monitoring of dynamic and heterogeneous cloud resources is essential to the overall operation of the cloud. In this paper, we review the principles of pervasive cloud monitoring, and discuss the requirements of a pervasive monitoring solution needed to support proactive and autonomous management of cloud resources. This paper reviews existing monitoring solutions used by the industry and assesses their suitability to support pervasive monitoring. We find that the

collectd

daemon is a good candidate to form the basis of a lightweight monitoring agent that supports high resolution probing, but it needs to be supplemented by high-level interaction capabilities for pervasive monitoring.

Gokce Gorbil, David Garcia Perez, Eduardo Huedo Cuesta

Source Model of TCP Traffic in LTE Networks

We propose a generator that represents traffic transmitted in mobile wireless networks. It is based on measurements of IP flows in a real, large-scale LTE network. The proposed tool generates TCP flow sizes and durations. We verify the generator by comparison with the traces available in the literature, and we propose application in discrete event simulators.

Paweł Foremski, Michał Gorawski, Krzysztof Grochla

A Few Investigations of Long-Range Dependence in Network Traffic

The paper presents measurements and analysis of a LAN long-range dependence traffic collected in IITiS PAN (The Institute of Theoretical and Applied Informatics of the Polish Academy of Sciences). Several methods of Hurst parameter estimation were used, the results obtained by the methods differ substantially. The analysis was made for the whole traffic and traffics generated by particular types of protocols. We seek for a dependence of Hurst parameter on a protocol type. Then, a MMPP (Markov-Modulated Poisson Process) model was applied to mimic the traces. It allows us to consider Markovian queueing models with long-range dependent and self-similar traffic, an important factor as we dispose an efficient software tool to solve numerically very large continuous-time Markov chains.

Joanna Domańska, Adam Domańska, Tadeusz Czachórski

Open Architecture for Quality of Service Monitoring at a National Research and Education Network

The Portuguese National Research and Education Network (NREN) has a set of proprietary

appliances

for Quality of Service (QoS) monitoring probes within its backbone, very much dependant both on

hardware

and

software

details. Nowadays, several

open source

QoS monitoring systems and some Network Performance Measurement and Monitoring tools, developed both by GÉANT and Internet2 member institutions, are available. This work presents an open software architecture for generic hardware probes, based upon perfSONAR framework, for QoS monitoring and an associated management solution for software configuration and automatized distribution. This new architecture has been tested and deployed in a NREN backbone and QoS data has been gathered and integrated into the NREN’s database. This paper discusses this open source solution for NREN QoS monitoring, the clock synchronisation issues and, finally, discusses the results obtained in a real testbed deployment in the Portuguese NREN backbone.

Alexandre Santos, M. João Nicolau, Bruno Dias, Pedro Queiros

Security

Frontmatter

Mitigating for Signalling Attacks in UMTS Networks

The setup of connections in mobile UMTS network will trigger signalling messages between the mobile and the core network. Malicious mobile phone or defective applications can therefore trigger

Signalling Attacks

which result in excessive wireless bandwidth utilisation and workload for the control plane and core network. We overview the cause of these attacks and identify the parameters which play a role.

Mihajlo Pavloski, Erol Gelenbe

Cryptanalysis of a Cryptographic Algorithm that Utilizes Chaotic Neural Networks

This paper deals with the security and efficiency issues of a cryptographic algorithm which utilizes the principles of Chaotic Neural Network (CNN). The algorithm that we consider is the Delayed CNN-Based Encryption (DCBE), which is an encryption algorithm based on the Delayed CNN. Although the cryptographic algorithm has its own salient characteristics, our analysis show that, unfortunately, the DCBE is not secure since it is not capable of resisting known-plaintext, chosen-plaintext, and chosen-ciphertext attacks. Furthermore, unfortunately, the scheme is not efficient either, because of the large number of iteration steps involved in its implementation.

Ke Qin, B. J. Oommen

DroidCollector: A Honeyclient for Collecting and Classifying Android Applications

With the tremendous increase of Android malware, we need an automatic way of collecting Android applications and identifying the malware before they get installed on the end-user devices. In this paper, we propose a honeyclient for Android applications that will collect and classify Android applications. We first present an overview of the honeyclient. Then, we survey the different ways of infecting Android mobile devices which will shed the light on the honeyclient’s design. Finally, we describe every component of the honeyclient, namely a crawler to build a list of suspicious URLs, a client to visit the suspicious URLs, extract Android applications and analyze them, and a malware detector to classify the collected Android applications. We use a light version of the Android browser to visit the suspicious URLs enabling us to scale the visits up and an Android emulator to analyze the Android applications. As for the malware detector, we use a combination of misuse and anomaly detector allowing us to detect already known malware and new variants.

Laurent Delosières, Antonio Sánchez

Real Time Wireless Packet Monitoring with Raspberry Pi Sniffer

This paper proposes a real time wireless packet monitoring system using a Raspberry Pi. The system is a low cost alternative to commercial packet capture devices and analysis software. In our solution, captured packets from sniffer are sent to main server to gather statistics. Packets are analyzed and only the relevant data are stored in database. A notification server developed in Node.js provides communication between database and user interface developed with Django web framework. The performance of the proposed solution is successfully evaluated in an environment with multiple wireless networks. Results are presented.

Yusuf Turk, Onur Demir, Sezer Gören

Visual Analytics for Enhancing Supervised Attack Attribution in Mobile Networks

Researchers have recently uncovered numerous anomalies that affect 3G/4G networks, caused either by hardware failures, or by Denial of Service (DoS) attacks against core network components. Detection and attribution of these anomalies are of major importance for the mobile operators. In this respect, this paper presents a lightweight application, which aims at analyzing signaling activity in the mobile network. The proposed approach combines the advantages of anomaly detection and visualization, in order to efficiently enable the analyst to detect and to attribute anomalies. Specifically, an outlier-based anomaly detection technique is applied onto hourly statistics of multiple traffic variables, collected from one Home Location Register (HLR). The calculated anomaly scores are afterward visualized utilizing stacked graphs, in order to allow the analyst to have an overview of the signaling activity and detect time windows of significant change in their behavior. Afterward, the analyst can perform root cause analysis of suspicious time periods, utilizing graph representations, which illustrate the high-level topology of the mobile network and the cumulative signaling activity of each network component. Experimental demonstration on synthetically generated anomalies illustrates the efficiency of the proposed approach.

Stavros Papadopoulos, Vasilios Mavroudis, Anastasios Drosou, Dimitrios Tzovaras

Queuing Systems

Frontmatter

Quantum Queuing Networks Throughput Optimisation

We study models of quantum queues based on discrete time quantum walks with barriers. Our considerations refer to multi-servers queuing models. Input and output of jobs in the queue are realised by systems consisting of quantum coins and walkers. We show that presented models behave differently from the classical ones. We also present numerical methods for optimisation of jobs processed by the system. We simultaneously maximise jobs throughput in the system and minimise the number of jobs lost.

Dariusz Kurzyk, Piotr Gawron

A Queueing System with Probabilistic Inhomogeneous Vacations for Modeling Power-Saving in Wireless Systems with Retransmissions

In this paper, we investigate the power management of mobile devices using a variant of an M/G/1 queue with probabilistic inhomogeneous multiple vacations and generalized service process. Under the vacation scheme, at the end of a vacation the server goes on another vacation, with a different probability distribution, if during the previous vacation there have been no arrivals. The modified vacation policy depends on the initial vacation interval and the server selects randomly over

$$M$$

M

such vacation policies. The theoretical system can be applied for modeling the power saving mode of mobile devices in modern wireless systems. Moreover, the form of the service process properly describes the incremental redundancy retransmission scheme that provides different types of retransmissions in such systems. Steady state analysis is investigated, energy and performance metrics are obtained and used to provide numerical results that are also validated against simulations.

Ioannis Dimitriou

Stability Criterion of a General Multiserver Multiclass Queueing System

We consider a FCFS multiclass, multiserver queueing system with class-dependent and server-dependent service times. We find stability criterion of such a system using the regenerative approach. The key idea is to consider the basic queue-size process in the saturated system. Then we use renewal theory and a characterization of the limiting remaining regeneration time to establish that the basic regenerative process is positive recurrent. First we prove a sufficient condition, and then show that the system is unstable when this condition is violated. Some generalizations of this system are discussed as well.

Evsey Morozov

Performance Analysis of Cluster-Based Web System Using the QPN Models

The paper presents the cluster-based web system for that the time rate of changing a system offer is compared to the users’ interaction time with a system. Systems with rapidly changing offer are used in different domains including the electronic trading to build scalable distributed systems. In the paper, such systems are modeled by queueing Petri nets. The aim of this work was to develop simulation models of Internet system by queueing Petri net modeling environment tool to performance analysis. Developed models allow to evaluate their performance (e.g., response time). In our experimental analysis, we use clustered web environment with Apache DayTrader benchmark as an e-trading system. It helped to determine the parameters for the simulation models and then to verify the simulation results. The paper includes the selected results of models simulation. Our approach predicts the performance of the given application deploy on a selected platform.

Tomasz Rak

Traffic Engineering: Erlang and Engset Models Revisited with Diffusion Approximation

Classical Erlang and Engset formulae determining the availability of channels, loss probability, and characteristics of overflow traffic are still used in telecommunications. Moreover, they are also interesting for traffic management in mobile networks and in Internet. They are based on the assumption of Poisson flows and exponentially distributed time of connections. By means of diffusion approximation queuing models, we extend these results to the case of general distributions and transient state analysis.

Tadeusz Czachórski, Tomasz Nycz, Monika Nycz, Ferhan Pekergin

Data Classification and Processing

Frontmatter

CodeMagic: Semi-Automatic Assignment of ICD-10-AM Codes to Patient Records

In this study, we present a recommendation system for semiautomatic assignment of ICD-10-AM codes to free-text patient records. Only expert annotators can assign codes to medical texts, and the lack of standardization of medical documentation and language specific problems make the assignment process even more challenging. Our system assigns a set of top k ICD codes for each document by exploiting the idea of bag-of-words and by using Lucene search engine and Borda Count voting schema. Before the code assignment task, we preprocess patient records to form query bags. Experiments on a set of clinical records show that promising results are possible for semiautomatic assignment of ICD codes.

Damla Arifoğlu, Onur Deniz, Kemal Aleçakır, Meltem Yöndem

Utilizing Coverage Lists as a Pruning Mechanism for Concept Discovery

Inductive logic programming (ILP)-based concept discovery systems lack computational efficiency due to the evaluation of the large search spaces they build. One way to tackle this issue is employing pruning mechanisms. In this work, we propose a two-phase pruning mechanism for concept discovery systems that employ an Apriori-like refinement operator and evaluate the goodness of the concept descriptors based on their support value. The first step, which is novel in this work, is computationally inexpensive and prunes the search space based on the coverages of the concept descriptors. The second step employs a widely employed pruning mechanism based on the support value of the concept descriptors. The experimental results show that the first step leaves a search space reduced by 4–22 % to be evaluated by the second step, which is more costly.

Alev Mutlu, Abdullah Dogan, Pinar Karagoz

Context Sensitive Search Engine

In this paper, we use context information extracted from the documents in the collection to improve the performance of the search engine. In first step, we extract context using Lucene, DBPedia-Spotlight, and Wordnet. As the second step, we build a graph using extracted context information. In the third step, in order to group similar contexts, we cluster context graph. In the fourth step, we re-score results using context-clusters and context-information of documents, as well as queries. In the fifth step, we implement a data collection tool to collect gold-standard data. In the sixth and final step, we compare the results of our algorithm with gold-standard data set. According to the experimental results, using context information may improve the search engine performance but the collection should be relatively big.

Remzi Düzağaç, Olcay Taner Yıldız

A Formal Framework for Hypergraph-Based User Profiles

In this study, we propose a formal framework for user profile representation with hypergraphs. We exploit the framework to aggregate partial profiles of the individual to obtain a complete, multi-domain user model, since we aim to model the user from several perspectives. We use Freebase commons package concepts as predefined domains. The proposed user model is also capable of extracting

user domain capsules

, which models the user for the domain of interest. Moreover, using a hypergraph data structure results in solving connection-based problems easily, since the cost of local operations on a graph is low and independent of the size of the whole graph. Many problems in user modelling domain are connection-based problems, such as recommendation.

Hilal Tarakci, Nihan Kesim Cicekli

A Survey of Data Stream Processing Tools

In current international context boundaries set for applications are being pushed by the emergence of bursty and time-varying data streams required to be processed in near real-time. Furthermore, traditional techniques for data mining cannot be applied to data streams. Thus, stream-based applications must exhibit to excel at a plurality of requirements. According to defined rules presented in previous promulgated researches on this subject we differ stream-based applications and evaluate their aptitude to stream sources management. By this work we intend to present features and drawbacks of existing software coming from both industry and academic world, along with outlining our contribution to this field.

Marcin Gorawski, Anna Gorawska, Krzysztof Pasterak

Distributed RDFS Reasoning with MapReduce

We live in big data age in which many computational tasks either generate or need to use large datasets. This makes parallel and distributed computing a key for scalability. MapReduce is a programming model for processing large datasets in parallel and distributed fashion on cluster of computers. Today, since the size and complexity of RDFS documents increase rapidly, RDFS reasoning problem has to embrace and address the big data solutions. The output of RDFS reasoning job can be input to another job and the output of RDFS reasoning jobs grow big as the input documents gets bigger. In this study, an indexing method is proposed to speed up the RDFS reasoning over Hadoop clusters. We also explore the utility of caching and Hadoop ecosystem tools Apache Hive and Apache Pig for this task. Experimental evaluations on Dbpedia and Freebase datasets show that the indexing method is quite effective and offers scalable solutions. Performance of caching and Apache Hive is found acceptable too.

Yigit Cetin, Osman Abul

Performance Evaluation

Frontmatter

On-Demand Prefetching Heuristic Policies: A Performance Evaluation

Prefetching is a basic mechanism in the World Wide Web that speculates on the future behaviour of users to avoid the response delays. The relatively new requirement of the instantaneous response in some interactive services like On-Demand applications fuelled the need for ways to represent and reason about the challenging problem of prefetching control and performance evaluation. We study this challenging problem under a network protocol that adopts the simultaneous prefetching with equal-shared bandwidth, and in prefetching situations in which the controller seeks to reach a Zero-Cost system state as quickly as possible. Within this context, our first contribution is providing the backbone of a new paradigm for the performance evaluation of the On-demand prefetching policy. This backbone consists of our previously developed prefetching control model; the PREF-CT model and our previously developed optimal control algorithms; the ONE-PASS and the TREE-DEC algorithms. Our second contribution is developing the prefetching heuristic algorithm: the RBP. Compared to the optimal prefetching policies, the prefetching policies computed by our heuristic algorithm the RBP show significant performance in terms of the user’s latency and the bandwidth utilization.

Olivia Morad, Alain Jean-Marie

An Improved Upper Bound for the Length of Preset Distinguishing Sequences of Distinguished Merging Finite State Machines

In an earlier work, we have studied a special class of FiniteState Machines (FSMs) called Distinguished Merging FSMs (DMFSMs) and showed that one can construct a Preset Distinguishing Sequence (PDS) for a DMFSM with

$$n$$

n

states,

$$p$$

p

input symbols, and

$$r$$

r

output symbols in time

$$O(n^4+pn^2)$$

O

(

n

4

+

p

n

2

)

of length no longer than

$$O(n^3)$$

O

(

n

3

)

. In this work, we improve the upper bound for the length of a PDS to

$$(n-1)^2$$

(

n

-

1

)

2

, and present an algorithm to construct such a PDS for a DMFSM in time

$$O(n^4+pn^2)$$

O

(

n

4

+

p

n

2

)

or in time

$$O(rn^3+pn^2)$$

O

(

r

n

3

+

p

n

2

)

.

Canan Güniçen, Kemal İnan, Uraz Cengiz Türker, Hüsnü Yenigün

Time Parallel Simulation for Dynamic Fault Trees

Dynamic Fault Trees (DFT) are a generalization of Fault Trees which allow the evaluation of the reliability of complex and redundant systems. We propose to analyze DFT by a new version of time-parallel simulation method we have recently introduced. This method takes into account the monotonicity of the sample-paths to derive upper and lower bounds of the paths which become tighter when we increase the simulation time. As some gates of the DFT are not monotone, we adapt our method.

T. H. Dao Thi, J. M. Fourneau, N. Pekergin, F. Quessette

On Performance Evaluation of Loss-Based Overload Control Mechanism in Signaling System with SIP Protocol

In this paper, we compare two approaches of deploying a loss-based overload control mechanism in a signaling system with the SIP protocol, hop-by-hop, and end-to-end. For deploying these mechanisms, we propose a simple, non-preemptive priority queueing scheme for a SIP server. In this scheme, overload control messages receive high priority while other signaling messages are served with a low priority. This approach allows effective support of performance of the loss-based overload control mechanism in the system. To get knowledge about advantages of the proposed solution, we compare both approaches (hop-by-hop and end-to-end) with a system without overload control. The measured performance metrics are related to average call setup time and goodput.

Halina Tarasiuk, Jan Rogowski

Analysis of Physical, Social and Biological Systems

Frontmatter

Improving Event Recognition Using Sparse PCA in the Context of London Twitter Data

Motivated by some of the recent work based on using sparse principal component analysis to analyse social media, we propose an improvement which involves altering the input data matrices by considering what relationships they represent. Accordingly, we confirm our result by using Twitter data from London in the year 2012 as a medium to demonstrate on. Various alterations are made to the data matrix obtained from this data and the resulting matrices are then passed through a sparse principal component analysis algorithm. The resulting outputs are then analysed and it is shown that indeed the results do differ, with one particular variation consistently outperforming the rest. Our results are especially of interest when the data to be analysed can be represented by a binary matrix of some sort, e.g. in document analysis.

Theo Pavlakou, Arta Babaee, Moez Draief

Modeling Structural Protein Interaction Networks for Betweenness Analysis

Protein–protein interactions are usually represented as interaction networks (graphs), where the proteins are represented as nodes and the connections between the interacting proteins are shown as edges. Proteins or interactions with high betweenness are considered as key connector members of the network. The interactions of a protein are dictated by its structure. In this study, we propose a new protein interaction network model taking structures of proteins into consideration. With this model, it is possible to reveal simultaneous and mutually exclusive interactions of a protein. Effect of mutually exclusive interactions on information flow in a network is studied with weighted edge betweenness analysis and it is observed that a total of 68 % of bottlenecks found in p53 pathway network differed from bottlenecks found via regular edge betweenness analysis. The new network model favored the proteins which have regulatory roles and take part in cell cycle and molecular functions like protein binding, transcription factor binding, and kinase activity.

Deniz Demircioğlu, Özlem Keskin, Attila Gursoy

Three-Dimensional Haptic Manipulator Controlled Game in the Treatment of Developmental Coordination Disorder

Computer games are more and more used for serious purposes not only for entertainment. In few last years, we can observe very fast market development of serious games. They are used to education, simulation, research, health, and therapy purposes. This article describes a game controlled by Phantom Omni device that could be a rehabilitation tool in the treatment of developmental coordination disorder. The game focuses on usage of the feedback generated by the haptic device.

Anna Wałach, Agnieszka Szczęsna

Parallel Biological In Silico Simulation

It is crucial to understand how biological systems work, in particular the metabolic pathways, if we want to be able to understand diseases. One of the author has developed HSIM, a simulator dedicated to the biochemical simulation of the reactions inside the compartments of a virtual cell. Another author is the leader in the development of

bobpp

a high level parallelization framework. In this article, we propose an important improvement of bobpp in order to run HSIM in parallel. This will allow us to simulate more complex models, involving more chemical species and reactions, leading to more realistic results using a smaller amount of computing time.

Patrick Amar, Muriel Baillieul, Dominique Barth, Bertrand LeCun, Franck Quessette, Sandrine Vial

An Analysis on Empirical Performance of SSD-Based RAID

In this paper, we measure the I/O performance of five filesystems—EXT4, XFS, BTRFS, NILFS2, and F2FS, with five storage configurations—single SSD, RAID 0, RAID 1, RAID 10, and RAID 5. We show that F2FS on RAID 0 and RAID 5 with eight SSDs outperforms EXT4 by 5 times and 50 times, respectively. We also make a case that RAID controller can be a significant bottleneck in building a RAID system with high speed SSDs.

Chanhyun Park, Seongjin Lee, Youjip Won

Evaluation of Fairness in Message Broker System Using Clustered Architecture and Mirrored Queues

The paper presents a performance evaluation of a message broker system in various high availability configurations. We verify different redundancy architectures against queuing system performance on the example of Rabbit MQ system. We discuss fairness issues and find that the replication of queues may lead to significant differences in the performance offered to clients connected to different nodes of messaging system. Basing on the analysis, we propose cluster architectures which provide fair allocation of resource to multiple clients, while maintaining the redundancy and high performance.

Maciej Rostanski, Krzysztof Grochla, Aleksander Seman

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise