Leonid B. Sokolinsky

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.

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.

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.

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.

Revised Pursuit Algorithm for Solving Non-stationary Linear Programming Problems on Modern Computing Clusters with Manycore Accelerators

This paper is devoted to the new edition of the parallel Pursuit algorithm proposed the authors in previous works. The Pursuit algorithm uses Fejer’s mappings for building pseudo-projection on polyhedron. The algorithm tracks changes in input data and corrects the calculation process. The previous edition of the algorithm assumed using a cube-shaped pursuit region with the number of K cells in one dimension. The total number of cells is \(K^n\), where n is the problem dimension. This resulted in high computational complexity of the algorithm. The new edition uses a cross-shaped pursuit region with one cross-bar per dimension. Such a region consists of only \(n(K-1)+1\) cells. The new algorithm is intended for cluster computing system with Xeon Phi processors.

Scalability Evaluation of NSLP Algorithm for Solving Non-Stationary Linear Programming Problems on Cluster Computing Systems

The paper is devoted to a scalability study of the NSLP algorithm for solving non-stationary high-dimension linear programming problem on the cluster computing systems. The analysis is based on the BSF model of parallel computations. The BSF model is a new parallel computation model designed on the basis of BSP and SPMD models. The brief descriptions of the NSLP algorithm and the BSF model are given. The NSLP algorithm implementation in the form of a BSF program is considered. On the basis of the BSF cost metric, the upper bound of the NSLP algorithm scalability is derived and its parallel efficiency is estimated. NSLP algorithm implementation using BSF skeleton is described. A comparison of scalability estimations obtained analytically and experimentally is provided.

Decomposition of natural join based on domain-interval fragmented column indices

The paper describes decomposition of natural join relational operator based on the column indices and domain-interval fragmentation. This decomposition admits parallel executing the resource-intensive relational operators without data transfers. All column index fragments are stored in main memory in compressed form to conserve space. During the parallel execution of relational operators, compressed index fragments are loaded on different processor cores. These cores unpack fragments, perform relational operator and compress fragments of partial result, which is a set of keys. Partial results are merged in the resulting set of keys. DBMS use the resulting set of keys for building the resulting table. Described approach allows efficient parallel query processing for very large databases on modern computing cluster systems with many-core accelerators. A prototype of the DBMS coprocessor system was implemented using this technique. The results of computational experiments are presented. These results confirm the efficiency of proposed approach.

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.

Join decomposition based on fragmented column indices

The paper is devoted to the issue of decomposition of the join relational operator with the aid of distributed column indices. Such decomposition allows one to utilize the modern manycore accelerators (GPU or Intel Xeon Phi) to speed up the query execution for very large databases. Column indices are the new kind of index structures, which exploits “key-value” technics. The paper describes themethods of column index fragmentation based on domain intervals. This technic allows organizing the parallel query processing without exchanges. All column index fragments are stored in main memory in compressed form to conserve space. This approach can be implemented as a coprocessor for relational database systems. The database coprocessor is able to perform resourceintensive operations much more faster than a conventional DBMS.

Component-based development of cloud applications: a case study of the Mjolnirr platform

The use of a component-oriented approach to the development of distributed applications can significantly extend the scalability of the software systems. In this article we describe the Mjolnirr platform, providing deployment of private cloud PaaS systems, based on the component-oriented approach. Any library or Java application can be implemented on the basis of the Mjolnirr platform as a service. From a developer perspective, an application on the basis of the Mjolnirr platform is a set of independent components, which communicate through a message passing interface. We will discuss an architecture and basic aspects of the implementation of the Mjolnirr platform, consider a problem of workflows scheduling, approaches to simulation of cloud platforms by means of the private cloud PaaS-systems simulation system. Also, we will discuss the results of tests of the platform.