All issues

A hybrid cluster implies the use of computational devices with radically different architectures. Usually, these are conventional CPU architecture (e.g. x86_64) and GPU architecture (e. g. NVIDIA CUDA). Creating and exploiting such a cluster requires some experience: in order to harness all computational power of the described system and get substantial speedup for computational tasks many factors should be taken into account. These factors consist of hardware characteristics (e.g. network infrastructure, a type of data storage, GPU architecture) as well as software stack (e.g. MPI implementation, GPGPU libraries). So, in order to run scientific applications GPU capabilities, software features, task size and other factors should be considered.

This report discusses opportunities and problems of hybrid computations. Some statistics from tests programs and applications runs will be demonstrated. The main focus of interest is open source applications (e. g. OpenFOAM) that support GPGPU (with some parts rewritten to use GPGPU directly or by replacing libraries).

There are several approaches to organize heterogeneous computations for different GPU architectures out of which CUDA library and OpenCL framework are compared. CUDA library is becoming quite typical for hybrid systems with NVIDIA cards, but OpenCL offers portability opportunities which can be a determinant factor when choosing framework for development. We also put emphasis on multi-GPU systems that are often used to build hybrid clusters. Calculations were performed on a hybrid cluster of SPbU computing center.

The principle of minimal differentiation, based on the Hotelling model, is well known in the economy. It is applicable to horizontal differentiated goods of almost any nature. The Hotelling approach to modeling competition of oligopolies corresponds to a modern description of monopolistic competition with increasing returns to scale and imperfect competition. We develop a modification of the Hotelling model that endows a firm with a non-market advantage, which is introduced alike the valence advantage known in problems of political economy. The nonmarket (valence) advantage can be interpreted as advertisement (brand awareness of firms). Problem statement. Consider two firms competing with prices and location. Homogeneous consumers vary with its location on a segment. They minimize their costs, which additively includes the price of the product and the distance from them to the product. The utility function is linear with respect to the price and quadratic with respect to the distance. It is also expected that one of the firms (for certainty, firm № 1) has a market advantage d. The consumers are assumed to take into account the sum of the distance to the product and the market advantage of firm 1. Thus, the strategy of the firms and the consumers depend on two parameters: the unit t of the transport costs and the non-market advantage d. I explore characteristics of the equilibrium in the model as a function of the non-market advantage for different fixed t. The aim of the research is to assess the impact of the non-market advantage on the equlibrium. We prove that the Nash equilibrium exists and it is unique under additive consumers' preferences de-pending on the square of the distance between consumers and firms. This equilibrium is ‘richer’ than that in the original Hotelling model. In particular, non-market advantage can be excessive and inefficient to use.

It is known that the sound speed in medium that contain highly compressible inclusions, e.g. air pores in an elastic medium or gas bubbles in the liquid may be significantly reduced compared to a homogeneous medium. Effective nonlinear parameter of medium, describing the manifestation of nonlinear effects, increases hundreds and thousands of times because of the large differences in the compressibility of the inclusions and the medium. Spatial change in the concentration of such inclusions leads to the variable local sound speed, which in turn calls the spatial-temporal redistribution of acoustic energy in the wave and the distortion of its temporal profiles and cross-section structure of bounded beams. In particular, focal areas can form. Under certain conditions, the sound channel is formed that provides waveguide propagation of acoustic signals in the medium with similar inclusions. Thus, it is possible to control spatial-temporal structure of acoustic waves with the introduction of highly compressible inclusions with a given spatial distribution and concentration. The aim of this work is to study the propagation of acoustic waves in a rubberlike material with non-uniform spatial air cavities. The main objective is the development of an adequate theory of such structurally inhomogeneous media, theory of propagation of nonlinear acoustic waves and beams in these media, the calculation of the acoustic fields and identify the communication parameters of the medium and inclusions with characteristics of propagating waves. In the work the evolutionary self-consistent equation with integro-differential term is obtained describing in the low-frequency approximation propagation of intense acoustic beams in a medium with highly compressible cavities. In this equation the secondary acoustic field is taken into account caused by the dynamics of the cavities oscillations. The method is developed to obtain exact analytical solutions for nonlinear acoustic field of the beam on its axis and to calculate the field in the focal areas. The obtained results are applied to theoretical modeling of a material with non-uniform distribution of strongly compressible inclusions.

When person`s diseases is result of bacterial infection, various characteristics of the organism are used for observation the course of the disease. Currently, one of these indicators is dynamics of cytokine concentrations are produced, mainly by cells of the immune system. There are many types of these low molecular weight proteins in human body and many species of animals. The study of cytokines is important for the interpretation of functional disorders of the body's immune system, assessment of the severity, monitoring the effectiveness of therapy, predicting of the course and outcome of treatment. Cytokine response of the body indicating characteristics of course of disease. For research regularities of such indication, experiments were conducted on laboratory mice. Experimental data are analyzed on the development of pneumonia and treatment with several drugs for bacterial infection of mice. As drugs used immunomodulatory drugs “Roncoleukin”, “Leikinferon” and “Tinrostim”. The data are presented by two types cytokines` concentration in lung tissue and animal blood. Multy-sided statistical ana non statistical analysis of the data allowed us to find common patterns of changes in the “cytokine profile” of the body and to link them with the properties of therapeutic preparations. The studies cytokine “Interleukin-10” (IL-10) and “Interferon Gamma” (IFN$\gamma$) in infected mice deviate from the normal level of infact animals indicating the development of the disease. Changes in cytokine concentrations in groups of treated mice are compared with those in a group of healthy (not infected) mice and a group of infected untreated mice. The comparison is made for groups of individuals, since the concentrations of cytokines are individual and differ significantly in different individuals. Under these conditions, only groups of individuals can indicate the regularities of the processes of the course of the disease. These groups of mice were being observed for two weeks. The dynamics of cytokine concentrations indicates characteristics of the disease course and efficiency of used therapeutic drugs. The effect of a medicinal product on organisms is monitored by the location of these groups of individuals in the space of cytokine concentrations. The Hausdorff distance between the sets of vectors of cytokine concentrations of individuals is used in this space. This is based on the Euclidean distance between the elements of these sets. It was found that the drug “Roncoleukin” and “Leukinferon” have a generally similar and different from the drug “Tinrostim” effect on the course of the disease.

Industrial imaging or “machine vision” is currently a key technology in many industries as it can be used to optimize various processes. The purpose of this work is to create a software and hardware complex for measuring the overall and weight characteristics of cargo based on an intelligent system using neural network identification methods that allow one to overcome the technological limitations of similar complexes implemented on ultrasonic and infrared measuring sensors. The complex to be developed will measure cargo without restrictions on the volume and weight characteristics of cargo to be tariffed and sorted within the framework of the warehouse complexes. The system will include an intelligent computer program that determines the volume and weight characteristics of cargo using the machine vision technology and an experimental sample of the stand for measuring the volume and weight of cargo.

We analyzed the solutions to similar problems. We noted that the disadvantages of the studied methods are very high requirements for the location of the camera, as well as the need for manual operations when calculating the dimensions, which cannot be automated without significant modifications. In the course of the work, we investigated various methods of object recognition in images to carry out subject filtering by the presence of cargo and measure its overall dimensions. We obtained satisfactory results when using cameras that combine both an optical method of image capture and infrared sensors. As a result of the work, we developed a computer program allowing one to capture a continuous stream from Intel RealSense video cameras with subsequent extraction of a three-dimensional object from the designated area and to calculate the overall dimensions of the object. At this stage, we analyzed computer vision techniques; developed an algorithm to implement the task of automatic measurement of goods using special cameras and the software allowing one to obtain the overall dimensions of objects in automatic mode.

Upon completion of the work, this development can be used as a ready-made solution for transport companies, logistics centers, warehouses of large industrial and commercial enterprises.

Industrial robots have made it possible for robotics to become a worldwide discipline both in economy and in science. However, their capabilities are limited, especially regarding contact tasks where it is required to regulate or at least limit contact forces. At one point, it was noticed that elasticity in the joint transmission, which was treated as a drawback previously, is actually helpful in this regard. This observation led to the introduction of elastic joint robots that are well-suited to contact tasks and cooperative behavior in particular, so they become more and more widespread nowadays. Many researchers try to implement such devices not with trivial series elastic actuators (SEA) but with more sophisticated variable stiffness actuators (VSA) that can regulate their own mechanical stiffness. All elastic actuators demonstrate shock robustness and safe interaction with external objects to some extent, but when stiffness may be varied, it provides additional benefits, e. g., in terms of energy efficiency and task adaptability. Here, we present a novel variable stiffness actuator with a magnetic coupler as an elastic element. Magnetic transmission is contactless and thus advantageous in terms of robustness to misalignment. In addition, the friction model of the transmission becomes less complex. It also has milder stiffness characteristic than typical mechanical nonlinear springs, moreover, the stiffness curve has a maximum after which it descends. Therefore, when this maximum torque is achieved, the coupler slips, and a new pair of poles defines the equilibrium position. As a result, the risk of damage is smaller for this design solution. The design of the joint is thoroughly described, along with its mathematical model. Finally, the control system is also proposed, and simulation tests confirm the design ideas.

The problem of optimal control of traffic flow in an urban road network is considered. The control is carried out by varying the duration of the working phases of traffic lights at controlled intersections. A description of the control system developed is given. The control system enables the use of three types of control: open-loop, feedback and manual. In feedback control, road infrastructure detectors, video cameras, inductive loop and radar detectors are used to determine the quantitative characteristics of current traffic flow state. The quantitative characteristics of the traffic flows are fed into a mathematical model of the traffic flow, implemented in the computer environment of an automatic traffic flow control system, in order to determine the moments for switching the working phases of the traffic lights. The model is a system of finite-difference recurrent equations and describes the change in traffic flow on each road section at each time step, based on retrived data on traffic flow characteristics in the network, capacity of maneuvers and flow distribution through alternative maneuvers at intersections. The model has scaling and aggregation properties. The structure of the model depends on the structure of the graph of the controlled road network. The number of nodes in the graph is equal to the number of road sections in the considered network. The simulation of traffic flow changes in real time makes it possible to optimally determine the duration of traffic light operating phases and to provide traffic flow control with feedback based on its current state. The system of automatic collection and processing of input data for the model is presented. In order to model the states of traffic flow in the network and to solve the problem of optimal traffic flow control, the CTraf software package has been developed, a brief description of which is given in the paper. An example of the solution of the optimal control problem of traffic flows on the basis of real data in the road network of Moscow is given.

Self-organization of molecules on a solid surface is one of the promising directions for materials generation with unique magnetic, electrical, and optical properties. They can be widely used in fields such as electronics, optoelectronics, catalysis, and biology. However, the structure and physicochemical properties of adsorbed molecules are influenced by many parameters that must be taken into account when studying the self-organization of molecules. Therefore, the experimental study of such materials is expensive, and quite often it is difficult for various reasons. In such situations, it is advisable to use the mathematical modeling. One of the parameters in the considered adsorption systems is the multiparticle interaction, which is often not taken into account in simulations due to the complexity of the calculations. In this paper, we evaluated the influence of multiparticle interactions on the total energy of the system using the transfer-matrix method and the Materials Studio software package. The model of monocentric adsorption with nearest interactions on a triangular lattice was taken as the basis. Phase diagrams in the ground state were constructed and a number of thermodynamic characteristics (coverage $\theta$, entropy $S$, susceptibility $\xi$) were calculated at nonzero temperatures. The formation of all four ordered structures (lattice gas with $\theta=0$, $(\sqrt{3} \times \sqrt{3}) R30^{\circ}$ with $\theta = \frac{1}{3}$, $(\sqrt{3} \times \sqrt{3})R^{*}30^{\circ}$ with $\theta = \frac{2}{3}$ and densest phase with $\theta = 1$) in a system with only pairwise interactions, and the absence of the phase  $(\sqrt{3}\times \sqrt{3}) R30^\circ$ when only three-body interactions are taken into account, were found. Using the example of an atomistic model of the trimesic acid adsorption layer by quantum mechanical methods we determined that in such a system the contribution of multiparticle interactions is 11.44% of the pair interactions energy. There are only quantitative differences at such values. The transition region from the  $(\sqrt{3} \times \sqrt{3}) R^{*}30^\circ$ to the densest phase shifts to the right by 38.25% at $\frac{\varepsilon}{RT} = 4$ and to the left by 23.46% at $\frac{\varepsilon}{RT} = −2$.

When determining therapeutic absorbed doses in the process of radioiodine therapy, the method of individual dosimetric planning is increasingly used in Russian medicine. However, for the successful implementation of this method, it is necessary to have appropriate software that allows modeling the pharmacokinetics of radioiodine in the patient’s body and calculate the necessary therapeutic activity of a radiopharmaceutical drug to achieve the planned therapeutic absorbed dose in the thyroid gland.

Purpose of the work: development of a software package for pharmacokinetic modeling and calculation of individual absorbed doses in radioiodine therapy based on a five-chamber model of radioiodine kinetics using two mathematical optimization methods. The work is based on the principles and methods of RFLP pharmacokinetics (chamber modeling). To find the minimum of the residual functional in identifying the values of the transport constants of the model, the Hook – Jeeves method and the simulated annealing method were used. Calculation of dosimetric characteristics and administered therapeutic activity is based on the method of calculating absorbed doses using the functions of radioiodine activity in the chambers found during modeling. To identify the parameters of the model, the results of radiometry of the thyroid gland and urine of patients with radioiodine introduced into the body were used.

A software package for modeling the kinetics of radioiodine during its oral intake has been developed. For patients with diffuse toxic goiter, the transport constants of the model were identified and individual pharmacokinetic and dosimetric characteristics (elimination half-lives, maximum thyroid activity and time to reach it, absorbed doses to critical organs and tissues, administered therapeutic activity) were calculated. The activity-time relationships for all cameras in the model are obtained and analyzed. A comparative analysis of the calculated pharmacokinetic and dosimetric characteristics calculated using two mathematical optimization methods was performed. Evaluation completed the stunning-effect and its contribution to the errors in calculating absorbed doses. From a comparative analysis of the pharmacokinetic and dosimetric characteristics calculated in the framework of two optimization methods, it follows that the use of a more complex mathematical method for simulating annealing in a software package does not lead to significant changes in the values of the characteristics compared to the simple Hook – Jeeves method. Errors in calculating absorbed doses in the framework of these mathematical optimization methods do not exceed the spread of absorbed dose values from the stunning-effect.

The article considers an approach to modeling the impact of sanctions and import substitution on the performance of high-tech product markets based on the use of control theory methods (operational calculus, z-transform). The model under consideration assumes that an equipment manufacturer supplies unique high-tech equipment to a high-tech product (HP) manufacturer that dominates the equipment consumer market. The HP manufacturer, fearing disruption of equipment supplies due to the introduction of all kinds of restrictions and sanctions, invests in the development of import-substituting equipment production in a third company, which can also find application in the external market, at the expense of deductions from its profits. The influence of the following factors and actions on the performance of the conditional market is analyzed: 1) the degree of inertia of the development and production development processes in the company; 2) the share of equipment of the import-substituting company supplied to the HP manufacturer; 3) sanctions (general and selective) on the supply of equipment to the company-manufacturer of the import substitution, as well as blocking the import substitution process in the third company by the first company.

The calculations show that the acceleration of the equipment development and production processes leads to a faster decrease in the production volumes of the first company. At the same time, an increase in price is observed, which is associated with a change in the parameters of the inverse demand function.

An increase in the share of equipment of the import-substituting company consumed by the second company can lead to a sharp increase in production volumes in the second and third companies, stabilization of production volumes in the first company and an increase in price.

The introduction of sanctions leads to a decrease in the production volumes and income of all companies relative to the baseline version. A significant change in price also occurs. However, due to the inertia of the equipment production processes in the example under consideration, a significant change in production volumes in the aggregate of companies occurs with a significant lag. This is especially characteristic of the third company, in which a noticeable deviation from the baseline version begins after 20 years. The blocking by the first equipment manufacturing company of investments in the development of import substitution in the third company ensures a relatively small gain for the first company in production volumes and NPV although allows to raise her market share.