Search

Search Results

• BIOINSPIRED SIMULATION METHOD FOR SCHEDULING OF PARALLEL FLOWS APPLICATIONS IN GRID-SYSTEMS

  D.Y. Kravchenko, Y.A. Kravchenko, V. V. Kureichik, A.E. Saak

  2020-07-20

  Abstract ▼

  The article is devoted to solving the problem of parallel requests scheduling flows in spatially
  distributed computing systems. The relevance of the task is justified by a significant increase in
  the demand for the distributed computing paradigm in the conditions of information overflow and
  uncertainty. The article discusses the problems of scheduling user requests that require severalprocessors at the same time, which goes beyond the classical theory of schedules. The aspects of
  the efficiency of using heuristic algorithms for scheduling planar resources are analyzed. The
  reasons for their insufficiency are determined both in terms of effectiveness and empirical approaches.
  The paper proposes to solve the problem of scheduling parallel applications based on
  the integrated application of intelligent agents coalition and an event simulation model. It is proposed
  to classify incoming applications on the basis of using a modified bio-inspired optimization
  method for cuckoo search. The joint use of a coalition of intelligent agents and a bio-inspired
  method will allow for unprecedented parallelism of calculations, and the subsequent determination
  of the processing classified applications ways on the basis of a simulation model will allow us
  to form sets of alternative solutions to speed up problem solving and optimize the distribution of
  available computing resources depending on the sets of incoming applications. To evaluate the
  effectiveness of the proposed approach, a software product was developed and experiments were
  conducted with a different number of incoming applications. Each incoming application has a
  certain set of attributes, which is a vector of the application characteristics. The degree of the
  application similarity feature vector and the vertex reference feature vector in the distributing
  simulation model is a classification criterion for the application. To improve the quality of the dispatch
  process, new procedures for duplicating unclassified applications have been introduced, which
  allow intensifying the search for matches in feature vectors. It also provides backup dispatching trajectories
  necessary for processing precedents for the appearance of applications with absolute priority
  at the inputs. The quantitative estimates obtained demonstrate time savings in solving problems of
  relatively large dimension (from 500,000 vertices) of at least 10%. The time complexity in the considered
  examples was O (n 2). The described studies have a high level of theoretical and practical significance
  and are directly related to the solution of classical problems of artificial intelligence
  aimed at finding hidden dependencies and patterns on a large set of big data.

• AN EVOLUTIONARY ALGORITHM FOR SOLVING THE DISPATCHING PROBLEM

  V. V. Kureichik, A. E. Saak, Vl.Vl. Kureichik

  2021-07-18

  Abstract ▼

  The paper considers one of the most important optimization tasks – the dispathing task that belongs
  to the class of NP-complex optimization problems. The paper presents the formulation of this
  problem. In Grid systems the array of users' requests for computer services is modelled by an extended
  linear polyhedral of coordinate resource rectangles. In this case, dispatching is represented
  by the localization of a linear polyhedron in the envelope of the area of computational and
  time resources of the system according to the multipurpose criterion of the quality of the applied
  assignment. Due to the complexity of this problem, the authors propose methods of evolutionary
  modelling for its effective solution and describe a modified evolutionary search architecture.
  Three additional blocks are introduced as a modification. This is a block of "external environment",
  a block of evolutionary adaptation and a block of "unpromising solutions." The authors
  have developed a modified evolutionary algorithm that uses the Darwin’s and Lamarck’s evolution
  models. This makes it possible to significantly reduce the time for obtaining the result, partially
  solve the problem of premature convergence of the algorithm, and obtain sets of quasi-optimal
  solutions in polynomial time. A software module has been developed in the C # language. A computational
  experiment has carried out on test examples and shown that the quality of solutions
  obtained on the basis of the developed evolutionary algorithm is, on average, 5 percent higher
  than the results of solutions obtained using the known algorithms of sequential, initial-ring and
  level at comparable time, which indicates the effectiveness of the proposed approach.

• CENTRAL-RING POLYNOMIAL ALGORITHM FOR DISTRIBUTION OF COMPUTATION-TIME RESOURCES IN GRID SYSTEMS

  D.Y. Kravchenko, Y.A. Kravchenko, E.V. Kuliev, A.E. Saak

  2022-08-09

  Abstract ▼

  The article is devoted to solving the problem of computational and time resources distribution
  in grid systems based on the adaptation of polynomial algorithms to quadratic types of user applications.
  The relevance of demand distribution validity problem for the distributed computing paradigm
  in the context of information redistribution and uncertainty. The article deals with the problems of
  scheduling heterogeneous computing resources in solving complex professional and scientific problems
  achieved at different points in time, based on identifying resources by significant manifestations
  of commitment and probability. A comparative review of consumption has been carried out. The
  statement of the problem to be solved in the chosen research area is formulated. The problem of
  scheduling a grid system with a centralized multiarchitecture, which uses the task solution of a
  group-site, is substantiated. The use of this architecture requires the development of heuristic algorithms
  for the distribution of computing resources, taking into account the properties of application
  arrays and assessing the schedule compliance. Eliminating the occurrence of scheduling errors requires
  the development of a formal apparatus that will identify the prospects of the application, introduce
  their typing and build heuristic algorithms with quality assessment, selected for certain types.
  The development of such a formal apparatus is an urgent task. An equally important task within the
  framework of this mechanism is the construction of resource parity models and interaction between
  users and the computing system models. The authors proposed to solve the problem of scheduling
  computing resources based on the development and study of polynomial scheduling algorithms for
  arrays of hyperbolic applications. The main theoretical accuracy of this study is the creation of a
  formal scheduling apparatus, including the definition of resource sugar, as a model of user applications,
  based on the performance of an operation in the scheduling environment on a set of resource
  muscles. The scientific novelty of the research lies in the development of a central-ring polynomial
  algorithm for the distribution of computational time resources in grid systems, which involves an
  automatic scheduling algorithm for computing systems, adaptation to quadratic types of user applications
  and improves the efficiency of computational time resources distribution. To evaluate the
  developed efficiency of the software application algorithm and the conducted computational experiment
  with rapidly generated classes of computational resources. Obtained comparative results of the
  proposed algorithm practical efficiency experimental studies for the distribution of computational
  and time resources. The described studies have a high level of theoretical and practical significance
  and are directly related to the solution of artificial intelligence classical problems.

• INTEGRATED MODEL FOR SOLVING THE PROBLEM OF REQUEST DISPATCHING

  А.E. Saak, L.A. Gladkov, N.V. Gladkova

  2023-02-17

  Abstract ▼

  The paper considers the problem of scheduling. The paradigm of organization of distributed
  computing based on Grid-computing is considered. The classification of task scheduling systems is
  given. Various approaches to solving the scheduling problem are described. A model of the task of
  servicing applications based on the principles of the theory of queuing systems is presented. The
  task statement is formulated on the basis of Grid-scheduling. The concept of a resource rectangle
  is proposed. The environment for scheduling resource rectangles is defined. A model is proposed
  that allows formalizing the user's request for service by the concept of a resource (non-Euclidean)
  rectangle. Instead of the principle of optimization based on the machine search for the best distribution
  of the array of resource rectangles, a heuristic principle was proposed, which made it possible
  to reduce the amount of necessary calculations. The proposed heuristic scheduling algorithm
  makes it possible to take into account the properties of the array and evaluate the quality of solutions.
  Models of the demand environment in the form of single cubic faces are constructed.
  The model of cubic faces is generalized to the experiment of cubic layers. The description of the
  demand model used is given. A model of the resource supply environment in the form of a canonical
  pyramid is constructed and the concept of a canonical demand-supply experiment for model
  homogeneous resource elements is introduced. A truncation of the supply-demand experiment has
  been introduced. A hybrid model based on a combination of evolutionary search principles and
  fuzzy control methods is proposed. To solve scheduling problems, it is proposed to use evolutionary
  algorithms. A modified solution coding technique and new modifications of genetic operators
  for solving scheduling problems have been developed. A block diagram of the algorithm for solving
  the problem under consideration is presented, taking into account the use of a fuzzy logic controller.
  Computer simulation has been performed and the results of computational experiments
  have been presented. The features of the proposed method are revealed, its advantages and disadvantages
  are formulated.

• BIOINSPIRED METHOD FOR CLASSIFICATION OF DISTRIBUTED RESOURCES FOR DISPATCHING IN GRID-COMPUTING

  D.Y. Kravchenko , Y.A. Kravchenko, V.V. Markov , A. E. Saak

  2021-11-14

  Abstract ▼

  The article is devoted to solving the problem of scheduling distributed computing resources
  based on their classification by the bioinspired search method to improve the efficiency of gridcomputing
  functioning. The relevance of the problem is justified by a significant increase in the
  demand for the paradigm of distributed computing in conditions of information overflow and uncertainty.
  The article deals with the problems of scheduling heterogeneous computing resources
  when solving complex professional and scientific problems arriving at different points in time,
  based on the classification according to significant signs of resource compliance and readiness. A
  comparative review of existing analogues is carried out. The formulation of the problem to be
  solved in the context of the selected research topic is formulated. The strategy of choosing
  bioinspired modeling for solving the problem has been substantiated. The aspects of various decentralized
  bioinspired methods effectiveness of the use are analyzed. It is proposed to solve the
  problem of scheduling computational resources based on determining the correspondence of the
  resource to the required class. The classification is carried out on the basis of the bioinspired
  optimization method application, built on the basis of the Fish School Search algorithm. The use of
  the population bioinspired method allows us to provide unprecedented parallelism in obtaining
  alternative solutions and to optimize the distribution of available computing resources depending
  on the sets of significant features. The object of the research is the processes of data classification,
  which include ordered sequences of actions aimed at the distribution of computing resources by
  classes of problems to be solved. The subject of the research is bioinspired methods for solving the
  problem of data classification in grid-computing. To evaluate the effectiveness of the proposed
  method, a software application was developed and a computational experiment was carried outwith a different number of computing resources generated classes. Each computing resource has a
  certain set of attributes, which is a vector of its features. The cosine measure of the similarity between
  a resource attributes vector and a certain class attributes vector is a classification criterion.
  To improve the quality of the dispatching process, the task of classifying computing resources is
  solved for a variety of options for organizing the flows of complex tasks to be solved in gridcomputing.
  The obtained quantitative estimates demonstrate the time savings in solving the problems
  of scheduling distributed computing resources based on their classification by the bioinspired
  search method at least 7 %. The time complexity in the considered examples was . The described
  studies have a high level of theoretical and practical significance and are directly related
  to the solution of artificial intelligence classical problems.