Search

During recent years, big data explosion and the increase in main memory capacity, on the one hand, and the need for faster data processing, on the other hand, have caused the development of various in-memory processing tools to manage and analyze data. Engaging the speed of the main memory and advantaging data locality, these tools can process a large amount of data with high performance. Apache Ignite, as a distributed in-memory platform, can process massive volumes of data in parallel. Currently, this platform is CPU-based and does not utilize the GPU’s processing resources. To address this concern, we introduce Ignite-GPU that uses the GPU’s massively parallel processing power. Ignite-GPU handles a number of challenges in integrating GPUs into Ignite and utilizes the GPU’s available resources. We have also identified and eliminated time-consuming overheads and used various GPU-specific optimization techniques to improve overall performance. Eventually, we have evaluated Ignite-GPU with the Genetic Algorithm, as a representative of data and compute-intensive algorithms, and gained more than thousands of times speedup in comparison with its CPU version.

Introduction: Nowadays, admission of students for medical specialty courses has faced some challenges. Due to the special importance of the duties and mission of physicians and the necessity of careful selection in their admission, the purpose of this study was to determine the contribution of academic performance and cognitive factors in students' admission to medical university. Methods: This was a descriptive cross-sectional study. The study population consisted of all the candidates of the specialty exams of Medical Sciences Universities in 2018, The members of these two groups were 1096 that score of them were analyzed in GPA, pre-internship score and cognitive test scores. The sample was selected by simple random sampling method. Research tools were California analytical thinking test, Watson Glazer critical thinking test, and reading comprehension test whose reliability by Cronbach's alpha coefficient came out as 0.79, 0.76, and 0.80, respectively. Data were analyzed using SPSS and R software and logistic regression, stepwise regression and independent t-test were used to answer the research questions. Result: The results showed that variables of GPA (t = 8.59, P = 0.0001), pre-internship test score (t = 7.45, P = 0.0001) and cognitive test score (t = 4.48, P = 0.0001) have a significant role in predicting candidates' performance in specialty entrance exam. Also, these variables could predict the status of admission or rejection of students in specialized courses (X2 = 630.63, P < 0.0001). The results of independent t-test showed that there was a significant difference between the performance of men and women in specialty entrance exam (t = 4.91, P = 0.0001), GPA (t = 4.67, P = 0.0001) and cognitive tests (t = -2.08, P = 0.038). Conclusion: Cognitive ability of medical university specialty candidates and their previous academic performance are the most important predictors of success in university entrance examination and admission. So it is necessary to accept students in specialized courses and universities based on essential competencies for medical students such as analytical thinking skills, logical thinking, problem solving and decision making, in addition to the entrance exams and performance of previous courses, in their assessments.

Multi-Party Non-Interactive Key Exchange (MP-NIKE) is a fundamental cryptographic primitive in which users register into a key generation centre and receive a public/private key pair each. After that, any subset of these users can compute a shared key without any interaction. Nowadays, IoT devices suffer from a high number and large size of messages exchanged in the Key Management Protocol (KMP). To overcome this, an MP-NIKE scheme can eliminate the airtime and latency of messages transferred between IoT devices. MP-NIKE schemes can be realized by using multilinear maps. There are several attempts for constructing multilinear maps based on indistinguishable obfuscation, lattices and the Chinese Remainder Theorem (CRT). Nevertheless, these schemes are inefficient in terms of computation cost and memory overhead. Besides, several attacks have been recently reported against CRT-based and lattice-based multilinear maps. There is only one modular exponentiation-based MP-NIKE scheme in the literature which has been claimed to be both secure and efficient. In this article, we present an attack on this scheme based on the Euclidean algorithm, in which two colluding users can obtain the shared key of any arbitrary subgroup of users. We also propose an efficient and secure MP-NIKE scheme. We show how our proposal is secure in the random oracle model assuming the hardness of the root extraction modulo a composite number., 14 pages, 3 figures

With the increasing rate of data generation in recent years, there is a need for modern tools to process these massive amounts of data. To that end, in-memory platforms are becoming increasingly popular, which can process high volumes of data at high speed and performance by utilizing Main Memory. Apache Ignite is one of the in-memory platforms that can process in parallel on multiple nodes. Although this platform provides many useful features, one of its limitations is the lack of utilizing the GPU's high processing power. Undoubtedly, using GPUs for operations that deal with heavy processing or high data volumes can be very beneficial, and significantly accelerate processing. One of the algorithms supported by Ignite is the Genetic Algorithm, which usually deals with large amounts of data, and might be very time-consuming. In this paper, we have provided an extension for Ignite in which users can utilize GPUs to run their Genetic Algorithm applications. Also, we have used various GPU-related optimization techniques to improve performance and finally evaluated our extension with three benchmarks. Our results proved the ease of use, and the high performance of the proposed work compared to Ignite.

Background: Impurities in narcotics are one of the factors that cause or exacerbate the symptoms of acute and chronic poisoning. Heavy metals as one of the impurities found in the opioids can cause toxicity to consumers. Opioids contamination and its complications are a social health problem in Middle Eastern countries such as Iran. Lead smugglers add more benefits during the drug production process in order to increase their weight. Objective: To determine the source of opioid contamination with lead metal. Materials and methods: Preparation of Opioid, Heroin, and Crock Afghan Drugs for quantitative analysis is used for Wet Digestion (oxidation or wet digestion), for analysis of a flame atomic spectrometer. Air-acetylene flame gas, halogen-based lamp based on the metal studied. Results: LOD and LOQ for lead, 2 PPM and 6 PPM spectroscopy were determined for flame atomic absorption spectrometry. Conclusion: According to the results, the level of lead in the poppy basin is negative as a source of opium production. However, lead was found in opium, burned opium, cracker Afghani and heroin as opium products. Which suggests that lead in the production of opium and its natural and industrial opioids is likely to be added to the benefits of smugglers and vendors.

Background: Methadone abuse in drug addiction centers and increased methadone consumption among addicts is a new cause of mortality due to substance abuse in forensic medicine. The usual method of extracting urine from corpses in most legal centers is a long and costly method of liquid-liquid extraction, which in addition to the high consumption of solvents and environmental contamination causes severe damage to the column of laboratory analysis devices. As a result, QuEChERS extraction method was proposed as an easy, fast, cheap and green way.

Building a research questionnaire and using assessment tools play a crucial role in social, management and behavioral sciences. Scales and questionnaires are deemed wrongly as the same tools, but in spite of the similarities that are basically and technichally different. Although, making a research scale or questionnaire may seem quite an easy job for son researchers, gathered results by these tools are normally miles away fro the reality. By this token, it necessitates some technical skills and scientif background. This article is to study the likert Scale, introduced by R. Likert in 193 as the most commonly used assessment tools and the most recent published researches' findings are explored to build and statically analy: the produced results developed by likert Scale.

The inrush current of transformer is one of the most important factors in the damaging of power transformers and reducing their useful life. Differential relays are one of the main and primary protections of power transformers. Therefore, the correct and fast operation of these devices is so important. In this paper, an inrush flow detection method is proposed in order to prevent the inaccurate performance of differential relay. In the proposed method, the analysis of the wavelet transform frequency time is used based on the extracted characteristics. Also, a criterion called correlation coefficient is used in order to create a suitable pattern and use it in the proposed algorithm. The simulation of the proposed algorithm is performed by the PSCAD software and implemented in the MATLAB environment. The results of the simulation confirm the accuracy of the proposed method.