Performance evaluation of the sparse matrix-vector multiplication on modern architectures

In this paper, we revisit the performance issues of the widely used sparse matrix-vector multiplication (SpMxV) kernel on modern microarchitectures. Previous scientific work reports a number of different factors that may significantly reduce performance. However, the interaction of these factors with the underlying architectural characteristics is not clearly understood, a fact that may lead to misguided, and thus unsuccessful attempts for optimization. In order to gain an insight into the details of SpMxV performance, we conduct a suite of experiments on a rich set of matrices for three different commodity hardware platforms. In addition, we investigate the parallel version of the kernel and report on the corresponding performance results and their relation to each architecture’s specific multithreaded configuration. Based on our experiments, we extract useful conclusions that can serve as guidelines for the optimization process of both single and multithreaded versions of the kernel.     

Performance evaluation of monitoring IoT systems using LoRaWan

The proliferation of smart devices, or even better, IoT devices, has led to the widespread development of applications that take advantage of these devices. Of particular interest is the precise localization of such a device. However, these use cases become extremely difficult when connectivity to end-devices is required even in areas where the signal is too low or different technologies co-exist for the transmission of the data. In this research work, we study LoRaWan and Wi-Fi as two possible candidates for data transmission. We are particularly focused on the study of the above technologies in terms of performance as well as application development that can be used as rescue monitoring systems. For this reason, we start by describing LoRa as an ideal low power and long-distance communication protocol on the IoT devices compared to the Wi-Fi network. We perform various simulations in terms of time on air transmission, bit error rate by changing important metrics to study the behavior of the whole mechanism. Based on our simulations, the main findings highlight that the contribution of a spreading factor and bandwidth optimizations can be applied to real hardware for real search and rescue (SAR) cases giving improved results in case of coverage and battery extension applications. As a continuation of our research, we developed a monitor application that collects and visualizes data from end-nodes (wearables). These data are processed gateway and network server to The Things Network (TTN) for further analysis. The proposed solution can be used in different rescue monitor scenarios such as identifying and find individuals of vulnerable groups or those belonging to group of people with a high probability of being lost. The purpose of the above solution is to overcome monitor problems on SAR cases, compare with WiFi and suggest a module supporting both technologies in order to be used in real experiments.

     

A case study on bypass testing of web applications

Society’s increasing reliance on services provided by web applications places a high demand on their reliability. The flow of control through web applications heavily depends on user inputs and interactions, so user inputs should be thoroughly validated before being passed to the back-end software. Although several techniques are used to validate inputs on the client, users can easily bypass this validation and submit arbitrary data to the server. This can cause unexpected behavior, and even allow unauthorized access. A test technique called bypass testing intentionally sends invalid data to the server by bypassing client-side validation. This paper reports results from a comprehensive case study on 16 deployed, widely used, commercial web applications. As part of this project, the theory behind bypass testing was extended and an automated tool, AutoBypass, was built. The case study found failures in 14 of the 16 web applications tested, some significant. This study gives evidence that bypass testing is effective, has positive return on investment, and scales to real applications.     

Injectability and mechanical properties of magnesium phosphate cements

Up to now magnesium phosphate cements are mainly being utilized in wastewater treatment due to their adsorptive properties. Recently they also have been shown to have a high potential as degradable biocements for application as replacement materials for bone defects. In comparison to degradable calcium phosphate cements they have the advantage of setting at neutral pH, which is favorable in biological environment. In this study two parameters of the cement composition, namely powder-to-liquid ratio (PLR) and citrate content, were varied in order to optimize the injectability properties of the cement paste and the mechanical properties of the reaction product. These properties were determined by means of testing setting time and temperature, paste viscosity, and injectability as well as phase composition and compressive strength of the set cements. Best results were obtained, when the cements were prepared with a PLR of 2.5 and a binder liquid consisting of an aqueous solution of 3 mol/l diammonium hydrogen phosphate and 0.5 mol/l diammonium citrate.     

Encryption and watermarking for the secure distribution of copyrighted MPEG video on DVD

This paper presents a complete system for the secure distribution of a copyrighted MPEG-1/2 video stored on a DVD-ROM disc. A combined selective watermarking and encryption method that operates in the compressed MPEG domain is introduced. Watermarking resistant to a number of attacks is used for copyright protection. The video quality deteriorates significantly due to encryption, thus restraining unauthorized viewers from viewing it. The video can only be viewed using the developed Secure MPEG Player, which performs real-time decryption of the encrypted video. The decryption requires a secret key that is extracted from the DVD-ROM disc in a cryptographically secure manner.This work was supported by the EU IST Project "ASPIS"     

An elastic partial shape matching technique

We consider the problem of partial shape matching. We propose to transform shapes into sequences and utilize an algorithm that determines a subsequence of a target sequence that best matches a query. In the proposed algorithm we map the problem of the best matching subsequence to the problem of a cheapest path in a directed acyclic graph (DAG). The approach allows us to compute the optimal scale and translation of sequence values, which is a nontrivial problem in the case of subsequence matching. Our experimental results demonstrate that the proposed algorithm outperforms the commonly used techniques in retrieval accuracy.     

Coastal Aquifer Management Based on the Joint use of Density-Dependent and Sharp Interface Models

In pumping optimization of coastal aquifers, the evaluation of the objective function and constraints using density-dependent models is overwhelmed by complex and time-consuming numerical simulations. To address those cases where the available density-dependent model runs are very limited, due to excessive computational burden, an efficient optimization strategy is developed. The proposed methodology uses an efficient sharp interface model jointly with a complex density-dependent model in an evolutionary optimization algorithm. While most evaluations are based on the sharp interface model, the density-dependent model is selectively called to evaluate promising solutions and to improve the predictions of the sharp interface model through the adaptive modification of the saltwater-freshwater density ratio. The method is tested for pumping optimization problems in confined and unconfined coastal aquifers with multiple pumping wells. The optimal solutions are compared to those obtained by density-dependent as well as by sharp interface optimization alone. Under a very restrictive computational budget, the best feasible solution is attained in less than 25 density-dependent model runs for two optimization problems of 10 and 20 decision variables. The results indicate that this optimization method leads to good feasible solutions and that an improved estimation of optimal pumping rates can be achieved within a limited computational budget. The method could also stand as an efficient preliminary exploration of the optimal search space, to provide good feasible starting points for the implementation of more comprehensive methods of coastal aquifer management.

     

The acceptance and use of computer based assessment

The effective development of a computer based assessment (CBA) depends on students’ acceptance. The purpose of this study is to build a model that demonstrates the constructs that affect students’ behavioral intention to use a CBA. The proposed model, Computer Based Assessment Acceptance Model (CBAAM) is based on previous models of technology acceptance such as Technology Acceptance Model (TAM), Theory of Planned Behavior (TPB), and the Unified Theory of Acceptance and Usage of Technology (UTAUT). Constructs from previous models were used such as Perceived Usefulness, Perceived Ease of Use, Computer Self Efficacy, Social Influence, Facilitating Conditions and Perceived Playfulness. Additionally, two new variables, Content and Goal Expectancy, were added to the proposed research model. Data were collected from 173 participants in an introductory informatics course using a survey questionnaire. Partial Least Squares (PLS) was used to test the measurement and the structural model. Results indicate that Perceived Ease of Use and Perceived Playfulness have a direct effect on CBA use. Perceived Usefulness, Computer Self Efficacy, Social Influence, Facilitating Conditions, Content and Goal Expectancy have only indirect effects. These eight variables explain approximately 50% of the variance of Behavioural Intention.     

On the optimal,fair and channel-aware cognitive radio network reconfiguration

In this work, we focus on the Joint Channel Assignment and Routing (JCAR) problem in Cognitive Radio Networks (CRNs) and especially on the optimal reconfiguration of secondary networks under the presence of primary users. Secondary CRN users need to adapt their transmission channels promptly, while effectively limit additional or escalating system modifications triggered by the intertweaved primary user activity. Our approach takes into consideration the underlying spectrum switching dynamics and concurrently aims at a fair resource allocation among the active network flows. We take an optimization perspective and formulate the JCAR and network reconfiguration problems as mixed integer linear programs, addressing fairness concerns as well. We propose a heuristic approach which is based on a sequential reduced search space methodology, in order to obtain efficiently solutions of otherwise tough and demanding reconfiguration problems. The operation, effectiveness and performance of the proposed mechanisms are evaluated through analysis and simulations under various working conditions. The obtained numerical results indicate the benefits of the proposed schemes in terms of overhead performance and their scaling properties with respect to more realistic and thus demanding topologies.