Organization of Parallel Query Processing in Multiprocessor Database Machines with Hierarchical Architecture

The development of database systems with hierarchical hardware architecture is currently a perspective trend in the field of parallel database machines. Hierarchical architectures have been suggested with the aim to combine advantages of shared-nothing architectures and architectures with shared memory and disks. A commonly accepted way of construction of hierarchical systems is to combine shared-memory (shared-everything) clusters in a unique system without shared resources. However, such architectures cannot ensure data accessibility under hardware failures on the processor cluster level, which limits their use in systems with high fault-tolerance requirements. In this paper, an alternative approach to construction of hierarchical systems is suggested. In accordance with this approach, the systems is constructed as an assembly of processor clusters with shared disks, with each cluster being a two-level multiprocessor structure with a standard strongly connected topology of interprocessor connections. A stream model for organization of parallel query processing in systems with the hierarchical architecture suggested is described. This model has been implemented in a prototype parallel database management system Omega designed for Russian multiprocessor computational systems MBC-100/1000. Our experiments show that the total performance of the processor clusters in the Omega system is comparable with that of the processor clusters with shared resources even in the case of great data skew. At the same time, the clusters of the Omega system are capable of ensuring a higher degree of data availability compared to the clusters with shared-memory architectures.     

Parallel processing of very large databases using distributed column indexes

The development and investigation of efficient methods of parallel processing of very large databases using the columnar data representation designed for computer cluster is discussed. An approach that combines the advantages of relational and column-oriented DBMSs is proposed. A new type of distributed column indexes fragmented based on the domain-interval principle is introduced. The column indexes are auxiliary structures that are constantly stored in the distributed main memory of a computer cluster. To match the elements of a column index to the tuples of the original relation, surrogate keys are used. Resource hungry relational operations are performed on the corresponding column indexes rather than on the original relations of the database. As a result, a precomputation table is obtained. Using this table, the DBMS reconstructs the resulting relation. For basic relational operations on column indexes, methods for their parallel decomposition that do not require massive data exchanges between the processor nodes are proposed. This approach improves the class OLAP query performance by hundreds of times.     

Methods of resource management in problem-oriented computing environment

One of the important classes of computational problems is problem-oriented workflow applications executed in distributed computing environment. A problem-oriented workflow application can be represented by a directed graph whose vertices are tasks and arcs are data flows. For a problem-oriented workflow application, we can get a priori estimates of the task execution time and the amount of data to be transferred between the tasks. A distributed computing environment designed for the execution of such tasks in a certain subject domain is called problem-oriented environment. To efficiently use resources of the distributed computing environment, special scheduling algorithms are applied. Nowadays, a great number of such algorithms have been proposed. Some of them (like the DSC algorithm) take into account specific features of problem-oriented workflow applications. Others (like Min–Min algorithm) take into account many-core structure of nodes of the computational network. However, none of them takes into account both factors. In this paper, a mathematical model of problem-oriented computing environment is constructed, and a new problem-oriented scheduling (POS) algorithm is proposed. The POS algorithm takes into account both specifics of the problem-oriented jobs and multi-core structure of the computing system nodes. Results of computational experiments comparing the POS algorithm with other known scheduling algorithms are presented.     

Boundary Condition Effects in Buckling of “Soft” Core Sandwich Panels

The effects of boundary conditions on the critical load level and the corresponding deflection mode shape of sandwich panels with a “soft” core due to in-plane loads are presented. The study is conducted using a closed-form high-order linearized buckling analysis that includes the influence of the transverse flexibility of the core as well as of the localized effects on the overall sandwich panel behavior, and allows the use of different boundary conditions for the upper and lower skin at the same section. The panel construction is general and consists of two skins (not necessarily identical), metallic or composite-laminated symmetric, and a soft core made of foam or a low-strength honeycomb. The closed-form high-order analysis yields the general buckling behavior of the structure, which means that the solutions obtained allow for interaction between the skins and the core. The solutions are general and are not based on separation of the buckling response on several types of uncoupled buckling modes, such as overall buckling, skins wrinkling, etc., as commonly used in the literature. The numerical scheme consists of finite differences to approximate the governing equations of the closed-form high-order formulation and to transform the set of linearized governing differential equations into an eigenvalue problem that is solved using the deflated iterative Arnoldi procedure. The influence of a general type of boundary conditions, including different conditions throughout the height of the same section and nonidentical conditions at the upper and lower skin, as well as of the core properties, on the buckling behavior of the sandwich panels is considered. The discrepancy between the Timoshenko-Reissner model and the present formulation is discussed. In particular, a partial fixity phenomenon due to the existence of the pinned boundary conditions, i.e., simply supported conditions, at the upper and lower skins at the edge is demonstrated. It is shown that the core properties affect the buckling loads and the corresponding modes of the panel in such a way that the structures with identical boundary conditions but with different cores may undergo different types of buckling such as overall and local as well as interactive loss of stability. The effect of an edge concentrated moment, induced by a couple and exerted on the skins only is also studied.     

Simulation of hierarchical multiprocessor database systems

The paper is dedicated to issues concerning simulation and analysis of hierarchical multiprocessor systems oriented to database applications. Requirements for a parallel database system model are given. A survey and comparative analysis of known parallel database system models are presented. A new multiprocessor database system model is introduced. This model allows us to simulate and evaluate arbitrary hierarchical multiprocessor configurations in the context of the OLTP class database applications. Examples of using the database multiprocessor model for simulation study of multiprocessor database systems are presented.     

Consistent higher-order free vibration analysis of composite sandwich plates

The consistent higher-order dynamic formulation for foam-type (soft) core sandwich beams was extended to the case of composite sandwich plates. Eight dynamic governing equations and the corresponding boundary conditions were derived through the application of Hamilton’s principle. The extended formulation was applied to the free vibration analysis of soft-core and honeycomb-core sandwich plates with anti-symmetric and symmetric lay-ups. The vibration results for the thin and thick composite sandwich plates obtained using the extended formulation were consistent with the predictions of the higher order mixed layerwise theory for laminated and sandwich plates. To simplify the formulation for the case of symmetric sandwich plates, the general dynamic formulation was decoupled into two systems of equations representing symmetric and anti-symmetric vibrations. The numerical study demonstrates the importance of the present formulation for the prediction of higher mode vibration response of composite sandwich plates.     

Query processing in a DBMS for cluster systems

The paper is devoted to the problem of effective query execution in cluster-based systems. An original approach to data placement and replication on the nodes of a cluster system is presented. Based on this approach, a load balancing method for parallel query processing is developed. A method for parallel query execution in cluster systems based on the load balancing method is suggested. Results of computational experiments are presented, and analysis of efficiency of the proposed approaches is performed.     

Prebuckling optimal design of orthotropic variable thickness plates for inplane loading

This paper is concerned with the optimal design of orthotropic rectangular plates for shear and compressive buckling loads. Expressions for the total potential as series summations are derived for double cosine thickness varying plates and hybrid double sine thickness varying plates. An efficient Fortran coded program for analysis of simply supported plates that utilizes the Rayleigh-Ritz method until convergence has been developed. It incorporates an optimization module that reactivates analysis as required. Four types of orthotropic plates were optimized in a one parameter optimization search. Results indicate a 33% increase in capacity for shear loading and an 89% increase in capacity for compressive loading.Notation A matrix of generalized eigenvalue problem - a vector of buckled shape amplitudes - a plate dimension inx-direction - a _mn amplitude of buckled shape of plate - B matrix of generalized eigenvalue problem - B ₁ –B ₁₆ coefficients of cube of plate thickness - B ₁ –B ₉ coefficients of square of plate thickness - b plate dimension iny-direction - CCEF parameter of the FORTRAN program - COEF parameter of the FORTRAN program - COEFFICIENTS module of the FORTRAN program - D ₁,D _x,D _y,D _xy flexural rigidities for orthotropic plate - flexural rigidities for orthotropic plate per cubic thickness - DI1-DI16 modules of the FORTRAN program - DI1A-DI16A modules of the FORTRAN program - DIO1-DIO9 modules of the FORTRAN program - DIO1A-DIO9A modules of the FORTRAN program - E Young's modulus - EIGEN module of the FORTRAN program - F total potential energy of buckled plate - F _c total potential for double-cosine shape representation - F _s total potential for hybrid shape representation - FK1-FK16 modules of the FORTRAN program - FK1A-FK16A modules of the FORTRAN program - FKS1-FKS16 modules of the FORTRAN program - FKS1A-FKS16A modules of the FORTRAN program - t _avr average thickness of nonuniform plate - t(x, y) thickness varying withx andy - t ₁,t ₂,t ₃ shape coefficients - I ₁ –I ₁₆ types of integrals - J ₁ –J ₁₆ types of integrals - K ₁ –K ₅ numerical factors - L ₁ –L ₉ types of integrals - L(t) differential operator - M ₁ –M ₉ types of integrals - m number of waves inx-direction - MAIN module of the FORTRAN program - N _xy shearing force per unit distance in middle surface of plate - NWRK parameter of the FORTRAN program - n number of waves iny-direction - N _x normal force inx-direction in middle - OPTIMIZER surface of plate - OPTIMIZER module of the FORTRAN program - p integer - POSTMAN module of the FORTRAN program - q integer - S area of plate - T work of external forces - U strain energy of bending - V volume of plate - w lateral displacement of plate - {W} vector of derivatives (–w, _xx ; –w _yy ;w _xy ) - x coordinate direction - y coordinate direction - z coordinate direction - constant of proportionality - combination of integers - numerical factor - variation - numerical factor - strain energy density - Poisson's ratio On leave from the Technion-Israel Institute of Technology, Technion City, Haifa 32000, Israel     

On the nature of “dynamic” strength of brittle bodies

The enhanced strength observed for bodies under dynamic loading has been attributed to fracturing duration, i.e. the time interval between the commencement of fracture at the most heavily loaded point of the body and the instant of completion of fracture which is recorded by a sensor. The “dynamic” strength of brittle bodies is shown to be related to the crack formation process.     

Survey of Architectures of Parallel Database Systems

The paper is devoted to the classification, design, and analysis of architectures of parallel database systems. A formalization of the notion parallel database system is suggested, which relies on a concept of a virtual machine. Based on this formalization, a new approach to the classification of architectures of parallel database systems is suggested. Requirements to parallel database systems are formulated, which serve as criteria for comparing various architectures. Various classes of architectures of parallel database systems are considered and compared.