Dynamic Control of Computation Consistency in the Parallel Dataflow Computing System

When developing high-performance multiprocessor computing systems, much attention is paid to ensuring uninterrupted operation, both in terms of hardware and software. In traditional computing systems, software is the main focus in address­ing these issues. The article discusses the solution to the issue of ensuring uninterrupted operation for the parallel dataflow computing system (PDCS), which implements the dataflow computational model with a dynamically formed context. Due to the features of the PDCS, it is proposed to implement this type of control in hardware, which will increase its efficiency, since the computational process will be controlled in dynamics, and not only in statics.

Ensuring Integrated Security as Part of Building Digital Architecture for Energy Companies

Today, there are models for building an enterprise architecture (EA) in general and power engineering in particular. However, these models provide only a systematic approach to the formation of EA, but do not contain metrics, therefore, they do not allow comparing different versions of architectures, and even more so to build an optimal EA. In our works, we have proposed a technology for designing an optimal EA. The article is the development of this technology and the integration of an approach to solving complex safety issues in an accident. This technology can be applied to any enterprise using information and digital technologies.

An Efficient Algorithm for Solving the Applied Problem Scheduling Optimization of a Parallel-Sequential System

The problem of optimal regulation of airline fleet schedules by reassigning aircraft to flights is considered. The optimal regulation of schedules is to create or change them in such a way that minimizes system losses due to current violations. As an estimate of losses, the total deviation of the adjusted schedule from the spetified departure schedules of aircraft is used. It is shown that the described technological system belongs to the category of parallel-sequential systems. Accordingly, the considered system control problem is NP-hard and does not have effective algorithms for exact solution. A brief overview of approaches to solving its modifications and related fleet management tasks is given. The original formal formulation is given, the decomposition of the problem is justified, and an algorithm for its approximate solution is presented. An illustrative example is given and comparative statistics of testing software implementations of the decomposition algorithm of the schedule control problem on real data are reflected, proving the actual effectiveness of the developed tools.

Support for Computing in Distributed Environments Based on Continuous Integration

Nowadays, tools for designing scientific applications often do not implement the required continuous integration capabilities of the applied software. Therefore, such overheads as the application development time and experiment execution makespan are substantially increased. In this regard, we propose a new approach to developing scientific applications and carrying out experiments with them. It is based on applying continuous integration to both the applied and system software in developing distributed applied software packages with a modular architecture using the Orlando Tools framework. Within the proposed approach, we provide integrating the Orlando Tools subsystems with the GitLab system and automating the development of package modules. At the same time, Orlando Tools fully support constructing and testing problem-solving schemes (workflows) that combine package modules located on environment resources with different computational characteristics. To this end, Orlando Tools provides the necessary configuring and setting up of computational resources. The practical significance of our study is substantial reduction overheads needed to experiment fulfillments and increase of the resource use efficiency.

Identification and Control of Multidimensional Delayed Processes in Conditions of Incomplete Information

Problems of identification and control of multidimensional discrete-continuous processes with delay in conditions of incomplete information about the object are considered. In such conditions, the form of parametric equations for various channels of the object is absent due to the lack of a priori information. Moreover, multidimensional processes have stochastic dependences of the components of the vector of output variables. Under such conditions, the mathematical description of such processes leads to a system of implicit equations. Nonparametric identification and control algorithms for multidimensional systems are proposed. The main task of modeling such processes is to determine the predicted values of the output variables from the known input. Moreover, for implicit equations, it is only known that one or another output variable can depend on other input and output variables that determine the state of a multidimensional system. In this study, a nontrivial situation arises when solving a system of implicit equations under conditions when the dependences between the components of the output variables are unknown. The application of the parametric theory of identification in this case will not lead to success. One of the possible directions is the use of the theory of nonparametric systems. The main content of the work is the solution of the identification problem in the presence of dependencies of the output variables and then the solution of the control problem for such a process. Here you should pay attention to the fact that when determining the reference actions for a multidimensional system, it is first necessary to solve the system of reference actions, since it is not possible to choose arbitrarily setting influences from the range of definition of output variables. Computational eXperiments aimed at investigating the effectiveness of the proposed identification and control algorithms are presented.

Mathematical Model of the Delay in Communication Networks Based on QS with a Time Lag

In the mathematical modeling of modern computer networks, telecommunication networks, traffic flows, logistics and many others, the methods of queuing theory are widely used. In turn, in studies of queuing systems (QS) G/G/1 with arbitrary distribution laws of intervals between adjacent requirements of the incoming flow and their service time, the spectral decomposition method (MSD) of solving the Lindley integral equation is often used. This method is based on the search for zeros and poles of the constructed spectral decomposition in the form of some fractional-rational function using numerical methods to determine the roots of polynomials. In this case, the coefficients of the polynomial in the numerator of the expansion are expressed through the unknown parameters of the distribution laws used to describe the QS. In the case of teletraffic research, usually these unknown parameters of the distribution laws can be determined through the numerical characteristics of the intervals between traffic packets by the method of moments. The purpose of this article is to present a fundamentally new mathematical model of a system formed by two flows with distribution laws shifted to the right. This is possible only for those probability distribution laws whose density functions are Laplace transformable. The main advantages of such systems, let us call them time lag systems, are that they provide less queue latency compared to conventional systems, and that they extend the range of traffic parameters. The article presents the results obtained on the average delay of requests in the queue for a system with exponential and hyper-Erlang distributions, an algorithm for calculating the average delay and the results of computational experiments in the Mathcad package.

Computational Models of Bio Heuristics Based on Physical and Cognitive Processes (Review)

Computational models of bio heuristics based on physical and cognitive processes are presented. Data on such characteristics of bio heuristics (including evolutionary and swarm bio heuristics) are compared.) such as the rate of convergence, computational complexity, the required amount of memory, the configuration of the algorithm parameters, the difficulties of software implementation. The balance between the convergence rate of bio heuristics and the diversification of the search space for solutions to optimization problems is estimated. Experimental results are presented for the problem of placing Peco graphs in a lattice with the minimum total length of the graph edges.

Investigation of Hierarchical Fuzzy Inference Systems, when Obtaining Integral Estimates of the Analyzed Objects

The purpose of this paper is to study the patterns of the formation of output values in hierarchical systems offuzzy inference. Hierarchical fuzzy inference systems (HFIS) are used to aggregate heterogeneous parameters during the assessment of the state of various elements of complex systems. The use of HFIS allows avoiding the "curse" of the dimension associated with a strong increase in the number and complication of the structure of the production rule, which is characteristic of conventional fuzzy inference systems (FIS), which aggregate the results of interaction of different values of input variables in one knowledge base. As part of the research, numerical experiments were carried out to study the features of the formation of output patterns in HFIS, based on FIS using the Mamdani and Takagi-Sugeno algorithms. As a result of the experiment, it was shown that the output values of the studied HFIS tend to be grouped in the region of fixed values, and the output pattern itself acquires a stepwise character. The revealed property allows using HFIS to distribute the objects of the analyzed sample into groups of states. This property can be used to solve problems of distributing objects into groups in conditions when it is difficult to form a training sample for machine learning methods, but at the same time there is knowledge of the expert group about the features of the functioning of the object of research. Additionally, the paper investigates the features of the formation of output patterns depending on the parameters of the membership functions describing the input variables in HFIS, which are based on FIS using the Mamdani algorithm and HFIS, which are based on FIS using the Takagi-Sugeno algorithm.

Technology for Conversion of Computer Tomography Results into Three-Dimensional Models

The development relates to information technology and can be used to process medical images. This task is achieved by the fact that DICOM files containing the results of computed tomography of the human head are subject to correction. They change the color of the pixels at the transition points from black to gray or vice versa, and also replace the black pixel with white. As a result, the percentage of loss of information about small bones is reduced from 22-31 % to 3-5 %, and, therefore, the accuracy of the three-dimensional model of the facial skull of the human head is increased. All this contributes to a significant improvement in the quality of facial surgery.

A Service for Parallel Processing of Protected Information Queries in a Multi-Server Environment

The principles of organizing an Internet service designed to implement parallel processing of protected queries in distributed information systems that are oriented to work in complex network environments with many processing servers are considered. Distinctive feature of the service consists in a possibility to combine sequential ("pipelined") and parallel query processing in a multi-server environment. The service architecture is based on the concept of a "Protected message" corresponding to a container for electronic documents (information query or response) that can be provided with one or more electronic digital signatures. In contrast, for example, to the Web services technology, the described service is based not on the model of calling methods of remote objects, but on the message exchange model. In this case, this means that all service processing functions (methods) have the same strict specification: they receive an array of objects of the "Protected message" program class as a parameter and return an array of objects of the same class. In sequential processing, both arrays usually contain one "Protected message" object each. However, when using parallel processing, arrays can contain multiple elements that correspond to the results of processing by multiple software threads. These processing functions are grouped into one or more dynamic libraries, which are connected to the PMS server when it starts working (each library can be considered as a remote analogue of a Web service), and become available to clients.