Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524849
A. Ivannikov, N. Levchenko, A. Okunev, A. Stempkovsky, D. Zmejev
A high-performance parallel dataflow computing system that implements a dataflow computing model solves problems in the field of supercomputing, the main one of which is low real performance when parallelizing tasks for hundreds of thousands of computational cores. In this article, a new architecture for the parallel dataflow computing system is proposed - a global distributed associative environment. The concept of the computing system organization is described, which consists in dividing the system into associative computational cores, which in turn consist of a set of local associative elements. The new architecture demonstrates significant advantages over the previous architecture, primarily due to the increased reliability of the system and the simplification of programming.
{"title":"Global Distributed Associative Environment - Evolution of Parallel Dataflow Computing System “Buran”","authors":"A. Ivannikov, N. Levchenko, A. Okunev, A. Stempkovsky, D. Zmejev","doi":"10.1109/EWDTS.2018.8524849","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524849","url":null,"abstract":"A high-performance parallel dataflow computing system that implements a dataflow computing model solves problems in the field of supercomputing, the main one of which is low real performance when parallelizing tasks for hundreds of thousands of computational cores. In this article, a new architecture for the parallel dataflow computing system is proposed - a global distributed associative environment. The concept of the computing system organization is described, which consists in dividing the system into associative computational cores, which in turn consist of a set of local associative elements. The new architecture demonstrates significant advantages over the previous architecture, primarily due to the increased reliability of the system and the simplification of programming.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122950091","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524834
A. Ermachikhin, V. Litvinov, Y. Vorobyov, Aleksei Maslov
Several different algorithms to measure physical parameters (especially, current-voltage characteristics) under fixed and variable temperature are presented here. Specifically, these algorithms are intended to achieve the uniform density of experimental data over the used bias voltage interval independent on the measured characteristic representation. Model and experimental data, obtained by using these algorithms are shown. The method of outliers detection during a current- voltage characteristic measurement are shown as well. Application of the adaptive algorithms during scientific experiment allows achieving better results by variation of investigating according to the input data.
{"title":"Application of Adaptive Algorithms for Measuring Temperature Current-Voltage Characteristics of Electronic Elements","authors":"A. Ermachikhin, V. Litvinov, Y. Vorobyov, Aleksei Maslov","doi":"10.1109/EWDTS.2018.8524834","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524834","url":null,"abstract":"Several different algorithms to measure physical parameters (especially, current-voltage characteristics) under fixed and variable temperature are presented here. Specifically, these algorithms are intended to achieve the uniform density of experimental data over the used bias voltage interval independent on the measured characteristic representation. Model and experimental data, obtained by using these algorithms are shown. The method of outliers detection during a current- voltage characteristic measurement are shown as well. Application of the adaptive algorithms during scientific experiment allows achieving better results by variation of investigating according to the input data.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115953799","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524695
V. Sapozhnikov, D. Efanov, V. Sapozhnikov, D. Pivovarov
The authors of the article developed a new approach to organizing concurrent error detection of logical devices of automatics and computing machinery based on the method of Boolean complement to “one out of three” constant-weight code. The introduced approach gives freedom of choice of testing equipment architecture in contrast to the established approaches. The example of managing a concurrent error detection system was studied in the article. It is based on step-by-step shipping of values of test functions taking into account testability of XOR elements and a checker with the introduction of uncertainties for the purpose of complexity optimization of the functions. The method in question is multipurpose and makes it possible to solve the task of building a concurrent error detection system with a totally self-checking structure in all cases.
{"title":"Self-Checking Concurrent Error Detection System Design Based on Boolean Complement Method to “1 out of 3” Code with Hardware Cost Optimization","authors":"V. Sapozhnikov, D. Efanov, V. Sapozhnikov, D. Pivovarov","doi":"10.1109/EWDTS.2018.8524695","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524695","url":null,"abstract":"The authors of the article developed a new approach to organizing concurrent error detection of logical devices of automatics and computing machinery based on the method of Boolean complement to “one out of three” constant-weight code. The introduced approach gives freedom of choice of testing equipment architecture in contrast to the established approaches. The example of managing a concurrent error detection system was studied in the article. It is based on step-by-step shipping of values of test functions taking into account testability of XOR elements and a checker with the introduction of uncertainties for the purpose of complexity optimization of the functions. The method in question is multipurpose and makes it possible to solve the task of building a concurrent error detection system with a totally self-checking structure in all cases.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"605 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116379818","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524672
V. Khryashchev, V. Pavlov, A. Priorov, A. Ostrovskaya
The goal of given investigation is to develop deep learning and convolutional neural network methods for automatically extracting the locations of objects such as water resource, forest and urban areas from given aerial images. We show how deep neural networks implemented on modern GPUs can be used to efficiently learn highly discriminative image features. For deep learning on supercomputer NVIDIA DGX-1 we used the marked image database UrbanAtlas, which contains images of 21 classes. Images obtained from the Landsat-8 satellites are used for estimation of automatic object detection quality. Object detection on aerial images has found application at urban planning, forest management, climate modelling, etc.
{"title":"Deep Learning for Region Detection in High-Resolution Aerial Images","authors":"V. Khryashchev, V. Pavlov, A. Priorov, A. Ostrovskaya","doi":"10.1109/EWDTS.2018.8524672","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524672","url":null,"abstract":"The goal of given investigation is to develop deep learning and convolutional neural network methods for automatically extracting the locations of objects such as water resource, forest and urban areas from given aerial images. We show how deep neural networks implemented on modern GPUs can be used to efficiently learn highly discriminative image features. For deep learning on supercomputer NVIDIA DGX-1 we used the marked image database UrbanAtlas, which contains images of 21 classes. Images obtained from the Landsat-8 satellites are used for estimation of automatic object detection quality. Object detection on aerial images has found application at urban planning, forest management, climate modelling, etc.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114600346","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524804
E. Birialtsev, Olga Zhibrik, A. Gusenkov, Polina Gusenkova, Yana Palacheva
This paper discusses the approach to a semantic search of mathematical expressions which allows seeking formulae by textual definitions of their variables. The proposed method allows establishing relations between terms, variables, and formulae. This method was tested in two different search services with different preprocessing approaches. The first service uses full-text indexing and responds to a query, the second uses ontology concepts and establishes relations in advance. In both cases, the proposed method produced high-relevant results. The method is applicable for search in documents containing mathematical expressions regardless the preliminary markup approach.
{"title":"Formula Markup Algorithm for Search in Scientific Documents","authors":"E. Birialtsev, Olga Zhibrik, A. Gusenkov, Polina Gusenkova, Yana Palacheva","doi":"10.1109/EWDTS.2018.8524804","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524804","url":null,"abstract":"This paper discusses the approach to a semantic search of mathematical expressions which allows seeking formulae by textual definitions of their variables. The proposed method allows establishing relations between terms, variables, and formulae. This method was tested in two different search services with different preprocessing approaches. The first service uses full-text indexing and responds to a query, the second uses ontology concepts and establishes relations in advance. In both cases, the proposed method produced high-relevant results. The method is applicable for search in documents containing mathematical expressions regardless the preliminary markup approach.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114565799","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524661
A. Tikhomirov, E. Omelyanchuk, A. Semenova, A. Smirnov, A. Bakhtin
The paper focuses on the noise and interference immunity properties of the direct spread spectrum systems in order to find out the best receiver structure for the terrestrial mobile communications considering limitations in energy consumption, computing resources and weight and size characteristics of mobile device in decentralized network. A DSSS system with two suitable receiver structures suitable for the mobile communication device was simulated. The simulation results show that direct spread spectrum systems have high immunity to pulse interference and the receiver structure with narrowband filtering on radio frequency shows better immunity to narrowband interference. It was proposed that for the case of high level narrowband interference a DC voltage compensation before the detector should be performed. The DSSS system with suggested compensation block can operate with the interference level which is larger than signal level in radio channel proportional to spreading factor.
{"title":"Direct Sequence Spread Spectrum System Noise and Interference Immunity Analysis","authors":"A. Tikhomirov, E. Omelyanchuk, A. Semenova, A. Smirnov, A. Bakhtin","doi":"10.1109/EWDTS.2018.8524661","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524661","url":null,"abstract":"The paper focuses on the noise and interference immunity properties of the direct spread spectrum systems in order to find out the best receiver structure for the terrestrial mobile communications considering limitations in energy consumption, computing resources and weight and size characteristics of mobile device in decentralized network. A DSSS system with two suitable receiver structures suitable for the mobile communication device was simulated. The simulation results show that direct spread spectrum systems have high immunity to pulse interference and the receiver structure with narrowband filtering on radio frequency shows better immunity to narrowband interference. It was proposed that for the case of high level narrowband interference a DC voltage compensation before the detector should be performed. The DSSS system with suggested compensation block can operate with the interference level which is larger than signal level in radio channel proportional to spreading factor.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126723213","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524634
A. V. Antonyuk, V. Stenin
The CMOS logical element of matching contains the STG DICE memory cell with transistors divided into two groups together with transistors of the output combinational logic. A single nuclear particle generates charges along its track. The collection of an induced charge just one group of transistors does not lead to the failure of the logical state of the cell. The results of the simulation presents for the element of matching as a part of an associative memory. It shows the hardness to a fault of the cell up to 70 MeV×cm2/mg of the linear energy transfer on the tracks of single particles. Noise pulses generate mostly in the output logic at the mode when the output has to be equal “1”. The correct results of matching data in registers of cells, when a noise pulse is, get more reliable using the summing circuits on the combinational logic but not using match lines.
{"title":"The Fault Tolerant CMOS Logical Element of Matching for a Content-Addressable Memory","authors":"A. V. Antonyuk, V. Stenin","doi":"10.1109/EWDTS.2018.8524634","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524634","url":null,"abstract":"The CMOS logical element of matching contains the STG DICE memory cell with transistors divided into two groups together with transistors of the output combinational logic. A single nuclear particle generates charges along its track. The collection of an induced charge just one group of transistors does not lead to the failure of the logical state of the cell. The results of the simulation presents for the element of matching as a part of an associative memory. It shows the hardness to a fault of the cell up to 70 MeV×cm2/mg of the linear energy transfer on the tracks of single particles. Noise pulses generate mostly in the output logic at the mode when the output has to be equal “1”. The correct results of matching data in registers of cells, when a noise pulse is, get more reliable using the summing circuits on the combinational logic but not using match lines.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126778031","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524680
A. Belyi, G. Osadchy, D. Efanov, Dmitry Shestovitskiy
Modern engineering facilities are unique objects in a single instance with a complex architecture and no less complex technical features. Such objects include the St. Petersburg Arena stadium built in 2016 in Saint Petersburg City. In addition to all the architectural delights, the stadium is also unique due to the decision to erect a sliding roof in order to protect the field and fans from the wind, rain and snowfall that are common in the region. The roof is a complex structure that requires constant monitoring of its technical condition. The system of continuous monitoring of technical condition of the roof structure of St. Petersburg Arena stadium is described in this paper. The uniqueness of the design was predetermined by the careful analysis of the measuring controllers arrangement in the computer model of the diagnostic object. The work describes the object under test in general and the sliding roof apart, as well as the connection points of the measuring controllers sensors indicating the required number of them (strain gauge, inclinometer, deflection sensors, and a weather station are used). The peculiarity and novelty of the problem being solved is that there are only four stadiums with a sliding roof in the whole world, and it is the only such construction in the Russian Federation. The technical solutions presented by the authors allow the transfer of monitoring results to the city's situation center, which in turn increases safety level and reduces the emergencies risks at the diagnostic site.
{"title":"Implementation of the Continuous Monitoring System for Technical Condition of the St. Petersburg Arena Stadium Sliding Roof","authors":"A. Belyi, G. Osadchy, D. Efanov, Dmitry Shestovitskiy","doi":"10.1109/EWDTS.2018.8524680","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524680","url":null,"abstract":"Modern engineering facilities are unique objects in a single instance with a complex architecture and no less complex technical features. Such objects include the St. Petersburg Arena stadium built in 2016 in Saint Petersburg City. In addition to all the architectural delights, the stadium is also unique due to the decision to erect a sliding roof in order to protect the field and fans from the wind, rain and snowfall that are common in the region. The roof is a complex structure that requires constant monitoring of its technical condition. The system of continuous monitoring of technical condition of the roof structure of St. Petersburg Arena stadium is described in this paper. The uniqueness of the design was predetermined by the careful analysis of the measuring controllers arrangement in the computer model of the diagnostic object. The work describes the object under test in general and the sliding roof apart, as well as the connection points of the measuring controllers sensors indicating the required number of them (strain gauge, inclinometer, deflection sensors, and a weather station are used). The peculiarity and novelty of the problem being solved is that there are only four stadiums with a sliding roof in the whole world, and it is the only such construction in the Russian Federation. The technical solutions presented by the authors allow the transfer of monitoring results to the city's situation center, which in turn increases safety level and reduces the emergencies risks at the diagnostic site.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122267032","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524764
K. Yusupov, V. Kosarev, A. Akchurin, Bulat Nasyrtdinov, A. Starovoytov, E. Yachmeneva, A. Mukhametzyanov, M. Khamiev
The paper describes the architecture and technical characteristics of the developed high-resolution ultrasonic borehole imager. Using the examples of comparing ultrasonic data images of well logging with core photos, its shown resolution in detecting fine details of a well with an opening of ~ 1 mm and others.
{"title":"The High Resolution Ultrasonic Well Imager","authors":"K. Yusupov, V. Kosarev, A. Akchurin, Bulat Nasyrtdinov, A. Starovoytov, E. Yachmeneva, A. Mukhametzyanov, M. Khamiev","doi":"10.1109/EWDTS.2018.8524764","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524764","url":null,"abstract":"The paper describes the architecture and technical characteristics of the developed high-resolution ultrasonic borehole imager. Using the examples of comparing ultrasonic data images of well logging with core photos, its shown resolution in detecting fine details of a well with an opening of ~ 1 mm and others.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123147970","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}
Pub Date : 2018-09-01DOI: 10.1109/EWDTS.2018.8524628
S. Ostanin, V. Andreeva, N. Butorina, D. Tretyakov
The use of modern high-speed electronic devices in high-tech industries requires high reliability of these devices. External factors such as radiation, high temperature, etc., often lead to the appearance of socalled soft faults of transient or intermittent. Such faults do not cause irrevocable changes in the equipment, and their manifestation lasts a limited time, not more than one clock cycle as a rule. Nevertheless it can affect the correct operation of the device. The accumulation of even small delays of gates in highspeed circuit along the path from the circuit input to the output can lead to an incorrect output signal. Such faults are called path delay faults. In this paper, we propose a method of fault-tolerant FSM network design for path delay faults, based on the use of a self-checking FSM network and a FSM network that implements the basic functionality without additional properties.
{"title":"Fault-Tolerant Synchronous FSM Network Design for Path Delay Faults","authors":"S. Ostanin, V. Andreeva, N. Butorina, D. Tretyakov","doi":"10.1109/EWDTS.2018.8524628","DOIUrl":"https://doi.org/10.1109/EWDTS.2018.8524628","url":null,"abstract":"The use of modern high-speed electronic devices in high-tech industries requires high reliability of these devices. External factors such as radiation, high temperature, etc., often lead to the appearance of socalled soft faults of transient or intermittent. Such faults do not cause irrevocable changes in the equipment, and their manifestation lasts a limited time, not more than one clock cycle as a rule. Nevertheless it can affect the correct operation of the device. The accumulation of even small delays of gates in highspeed circuit along the path from the circuit input to the output can lead to an incorrect output signal. Such faults are called path delay faults. In this paper, we propose a method of fault-tolerant FSM network design for path delay faults, based on the use of a self-checking FSM network and a FSM network that implements the basic functionality without additional properties.","PeriodicalId":127240,"journal":{"name":"2018 IEEE East-West Design & Test Symposium (EWDTS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131221587","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：