TASK Quarterly   Scientific Bulletin of the Centre of Informatics - Tricity Academic Supercomputer & networK   ISSN 1428-6394




  • G.Bergmanski, S.Feliziani and J.Rybicki, Model Angular Distribution Functions in CA3, CA4 and CA6 Structural Units of Glassy Systems

    We have calculated model partial angular distribution functions (pADFs) in CA3, CA4 and CA6 structural units, ie. an equilateral triangle with three vertical anions, A, and a central cation, C, a regular tetrahedron with four vertical anions, A, and a central cation, C and a square bipyramid with six vertical anions, A, and a central cation, C. The model pADFs were calculated employing a simple Monte Carlo procedure: the ions were being shifted at random within 3D spheres of radius r with uniform probability density and the AAA, ACA and CAA angles were calculated for each random configuration. Repeating the calculation 108-109 times produced smooth probability densities for the angles' values. Conventional reference data so obtained can be applied to estimate the overall degree of deformation of the considered structural units in numerically simulated materials.

  • Y.Fu, Y.Cai and K.Wu, Sensitivity-based Stability Evaluation of Closed Piping Networks

    Aiming at a closed-loop water system of HVAC engineering, the authors put forward an evaluation method of systems' stability based on sensitivity. It has three evaluating indexes: α - the summation of flow changes in other subcircuits influenced by resistance change in a certain subcircuit; β - the summation of flow changes in a certain subcircuit influenced by resistance changes in other subcircuits; γ - average β (or α) value of each subcircuit, reflecting the strength of regulating interference between subcircuits. The method is used to analyze the stability of a reverse return system (RRS) and a direct return system (DRS). The DRS subcircuit farthest from the heat source and the middle RRS are the least stable. Stability of the whole RRS is inferior to that of the DRS.

  • L.Pasini and S.Feliziani, Queue-based Simulators of P2P Networks

    A technique is introduced for the construction of simulators of telecommunication networks based on the P2P protocol with a decentralized directory. The context of the application concerns P2P networks used both by sharing software and by telecommunication applications, as in VoIP applications. P2P systems have the advantages of boundless scalability and the break-up insensibility of devices, but suffer from heavy traffic of signals and network delays. Therefore, simulation of signal traffic in P2P networks is an important issue. The technique we have developed enables individuation of a generic P2P network's components and their functional characterization. We have worked at this level of description in order to determine the architecture of every component as a model of queues' networks. We have therefore defined a library of new types of objects for the construction of P2P network simulators with the Qnap2.V9 programming language. In this context, we have defined a procedure automatically generating the simulator of an any P2P network. During its execution, the procedure reads data from a file containing a description of the structure of the considered P2P network and values of the characteristic parameters of its components. Results related to the simulation of three configurations of a P2P network are introduced in Section 5. They differ according to the technology used to realize the local interconnection networks. The context of the present paper is a research program developed at the University of Camerino in defining specific computer procedures aimed at construction of simulators of telecommunication and computing networks. This technique also enables simulation of the software elaboration phases on computer networks and evaluation of the impact of specific choices concerning the structure of transmission protocols and operation paradigms of software applications.

  • W.Oniszczuk, Analytical Investigation of Congestion-Avoidance Strategies in Closed-type Queuing Models of Computer Networks with Priority Scheduling

    A new approach is presented to modelling intelligent admission control and congestion avoiding mechanism, without rejecting new requests, embedded into a priority closed computer network. Most Call Admission Control (CAC) algorithms treat every request uniformly and hence optimize network performance by maximizing the number of admitted and served requests. In practice, requests have various levels of importance to the network, for example priority classes. Here, the investigated closed network with priority scheduling has been reduced to two service centres, which allows for decomposition of a larger network into a chain of individual queues, where each queue can be studied in isolation. A new algorithm (approach) of this special type of closed priority queuing systems is presented, including a node consisting of several priority sources generating tasks, designated as an Infinite Server (IS), and a service centre with a single service line. This model type is frequently described as a finite source, pre-emptive-resume priority queue (with general distribution of service time). The pre-emptive service discipline allows a task of lower priority to be returned to the head of a queue when a new task of higher priority arrives. A mathematical model of provisioning and admission control mechanism is also described. The idea behind this mechanism has been derived from the Hidden Markov Model (HMM) theory. It is crucial in the CAC process that the network manager obtains correct information about the traffic characteristics declared by the user. Otherwise, the quality of service (QoS) may be dramatically reduced by accepting tasks based on erroneous traffic descriptors. Numerical results illustrate the strategy's effectiveness in avoiding congestion problems.

  • R.Rzadkowski and V.Gnesin, The Influence of Steam Extraction Parameters on Unsteady Rotor Forces

    Aerodynamic unsteady forces of rotor blades in a turbine stage with steam extraction were calculated and blade failure was reported. A numerical calculation of the 3D transonic flow of an ideal gas through turbomachinery blade rows moving one in relation to another is presented. An ideal gas flow through mutually moving stator and rotor blades with periodicity on the whole annulus is described with unsteady Euler conservation equations, integrated using the explicit monotonous finite-volume difference scheme of Godunov-Kolgan and a moving hybrid H-H grid. In order to find the pressure distribution of steam parameters behind the rotor blades, calculations were performed for a steady flow through the stage and the steam extraction channels using the SPARC program and the unsteady forces of the rotor blade were calculated for four steam extraction conditions. For the maximal steam extraction the 10th harmonic of the axial force and moment was found to be close to the first natural frequency of the bladed disc with one nodal diameter. Thus, steam extraction can be the cause of the blade's failure.

  • R.Ogrodowczyk and K.Murawski, Numerical Simulations of Slow Magnetosonic Standing Waves in a Two-dimensional Hot Solar Coronal Slab

    We consider impulsive excitation of slow magnetosonic standing waves in a two-dimensional hot solar coronal slab. Results of numerical simulations have revealed that initially launched pulses trigger mainly fundamental slow mode and its first harmonic, depending on the pulses' spatial location. Our parametric study has shown these slow standing waves to exhibit wave periods of about 13min. The slow standing waves are accompanied by fast modes simultaneously present in the system.

  • P.Wroblewski, M.Kopec and K.Boryczko, SPH - a Comparison of Neighbor Search Methods Based on Constant Number of Neighbors and Constant Cut-off Radius

    Two methods of neighbor search for the SPH algorithm are presented, based on a constant number of neighbors and a constant cut-off radius. First, feasible methods of comparison were analyzed. Then, the two selected methods were compared visually and computationally. Considering the use of the SPH algorithm for simulating incompressible fluids, the obtained results suggest that the method with a constant cut-off radius is better than that with a constant number of neighbors. The simulation results of both methods are practically indistinguishable, while the computational costs favor one of them.

  • J.Dziedzic, Obtaining the Ionic Forces from the Total-energy Tight-binding Method

    Applying a non-orthogonal tight-binding method to calculate the ionic forces in a molecular-dynamics simulation vastly improves the transferability of the model to different environments, compared to traditional empirical potential-driven molecular-dynamics. In this paper we present the details of computing the derivatives of Hamiltonian and overlap matrix elements appearing in the Hellmann-Feynman expression for the ionic forces in the NRL-TB model of tight-binding. The validity of the presented expressions is supported by the results obtained using a tight-binding-driven molecular-dynamics program.