{"title":"The Queue State for Multiprocessor System with Waiting Time Restriction","authors":"V. Sahakyan, Artur Vardanyan","doi":"10.1109/CSITechnol.2019.8895093","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895093","url":null,"abstract":"The queue state in multiprocessor computing systems is an actual problem for their use in the process of tasks scheduling. In this paper, a system of differential equations is obtained describing the probabilities of the system state at time t, which can be solved using numerical methods. However, the probabilities sought-for the system M|M|m|n with a constraint on the residence time can be obtained in terms of the Laplace- Stieltjes transform.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122226724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Educational Potential of Internet Communications","authors":"T. Beregovskaya, S. Grishaeva, G. Margarov","doi":"10.1109/CSITechnol.2019.8895099","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895099","url":null,"abstract":"The modern Internet environment has great potential for education, as the functionality of Internet communication allows to make the educational process more technical, practical, coordinated, expanding the possibilities of access to both teachers and students to high-tech content. The article considers the functions of Internet communication and provides data on the study of the implementation of these functions from the point of view of students of leading Moscow universities.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130599810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Solving Scheduling Problems with Randomized and Parallelized Brute-Force Approach","authors":"R. Davidrajuh, Chunming Rong","doi":"10.1109/CSITechnol.2019.8895104","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895104","url":null,"abstract":"Most of the scheduling problems are NP-hard problems. Thus, they do not have polynomial-time solutions. The literature review provides hundreds of methods and approaches to find polynomial-time near-optimal solutions. Most of these approaches are based on genetic algorithms. Genetic algorithms have the power of scanning most of the solution space, and they are not vulnerable to hill-climbing phenomena. However, as this paper shows, genetic algorithms cannot be used if the rate of production of healthy offspring is very low. Hence, this paper proposes a novel approach that is based on randomized brute-force and inspired by genetic algorithms. Also, the proposed approach uses parallel processing.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114726654","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Accurate Pressure Calculation Method for CMP Modeling Using Fourier Analysis","authors":"R. Ghulghazaryan, Suren Alaverdyan, D. Piliposyan","doi":"10.1109/CSITechnol.2019.8895113","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895113","url":null,"abstract":"Chemical-mechanical polishing/planarization (CMP) is one of the key processes used in electronic chip manufacturing. In CMP, a rotating wafer is pressed facedown onto a rotating polishing pad. A chemical “slurry” containing abrasive particles and chemical reagents is deposited on the pad during polishing, and flows between the wafer and the pad. The combined action of the polishing pad, abrasive particles, and chemical reagents results in material removal and planarization of the wafer surface.In recent years, modeling of the CMP process has become critical for detection of planarity defects in chips before manufacturing. One of the key parameters affecting the surface planarization is the pressure, with which the wafer is pressed against the pad. Calculation of the pressure distribution across the wafer surface is crucial for modeling the CMP process. This pressure calculation typically uses contact mechanics methods that include solving an integral equation using fast Fourier (FFT) and inverse fast Fourier (IFFT) transforms. However, the kernel of the integral equation has singularities that lead to numerical instability.In this paper, we propose a method to avoid numerical instabilities in pressure calculation by using an analytical expression for the Fourier transform of the kernel function.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116494607","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Unified Software Platform for Discrete Tomography Reconstruction Algorithms","authors":"A. Margaryan","doi":"10.1109/CSITechnol.2019.8895175","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895175","url":null,"abstract":"In this paper a unified software platform DTR (Discrete Tomography Reconstruction) is introduced created for discrete tomography reconstruction algorithms.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116999158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Hyper: Distributed Cloud Processing for Large-Scale Deep Learning Tasks","authors":"Davit Buniatyan","doi":"10.1109/CSITechnol.2019.8895139","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895139","url":null,"abstract":"Training and deploying deep learning models in real-world applications require processing large amounts of data. This is a challenging task when the amount of data grows to a hundred terabytes, or even, petabyte-scale. We introduce a hybrid distributed cloud framework with a unified view to multiple clouds and an on-premise infrastructure for processing tasks using both CPU and GPU compute instances at scale. The system implements a distributed file system and failure-tolerant task processing scheduler, independent of the language and Deep Learning framework used. It allows to utilize unstable cheap resources on the cloud to significantly reduce costs. We demonstrate the scalability of the framework on running pre-processing, distributed training, hyperparameter search and large-scale inference tasks utilizing 10,000 CPU cores and 300 GPU instances with overall processing power of 30 petaflops.12","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127883423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"L1 Adaptive Control of Quadcopters","authors":"O. Gasparyan, Haykanush G. Darbinyan","doi":"10.1109/CSITechnol.2019.8895217","DOIUrl":"https://doi.org/10.1109/CSITechnol.2019.8895217","url":null,"abstract":"The issue of developing a control system for compensating external disturbances acting on the quadcopter is considered. The proposed design procedure consists in introducing a decoupling compensator into the system and subsequent synthesis of an adaptive regulator that provides compensation of external disturbances. It is shown that by virtue of choosing the value of the adaptation gain one can reduce the control system error to any arbitrary small value. The proposed control architecture can be used for developing fault-tolerant control systems of multirotor copters.","PeriodicalId":414834,"journal":{"name":"2019 Computer Science and Information Technologies (CSIT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121642967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}