RESEARCH OF THE INFLUENCE OF A NETWORK INTERFACE ON THE EFFICIENCY OF MODULAR MULTIPROCESSOR SYSTEMS

The paper is devoted to the approach development related to methodology definition for evaluation of the modular multiprocessor computing systems efficiency. At the same time, the main attention is focused on the impact peculiarities on this network interface value. The formation analysis of the multiprocessor system network interface architecture and the basic modes of its operation have been analyzed. To evaluate the processes occurring in the system during the information flows transmission, the network system bandwidth and the switch throughput were compared; which allowed determining the preconditions for optimal components selection of the multiprocessor computing system network interface. The performed researches also allowed deducing analytical relations for determining the optimal number of system nodes with different functioning modes. The selected processors coherency coefficient, network interface and value of the computing area are deduced. The derived analytical relationships also showed that the optimal number of blades in a multiprocessor computing system, that provide its highest speed, decreases with increasing computing power of the processors included. It is shown that the network data interchange among the multiprocessor computing system nodes the more likely to impede the overall computation process; the less time will be spent directly on solving a specific problem.

RESEARCH OF COMPUTING EFFICIENCY IN MODULAR MULTIPROCESSOR SYSTEMS

The article is devoted to the research of efficiency of a multiprocessor computing system in solving problems aimed at expanding the computing area. The basic regularities concerning the time of solving the problem are revealed, depending on the change in the multiprocessor system calculations area. The research is aimed at determining the deceleration factor associated with the increase of the computing area of a multiprocessor system when compared with the computer version with an unlimited computing area. The analytical ratios are derived for determining the calculations deceleration coefficient. A stage of simulation for calculations of the deceleration factor was carried out to determine the regularities of its change, depending on the application of a particular computing platform. The revealed tendencies of such a change point to the need to reconcile the components of the network interface and computing capabilities of the chosen computing platform. The derived analytical relations were aimed at determining the optimal number of nodes of a multiprocessor system which allow the minimum delay of calculations.

GENE LOSS PREDICTION BASED ON GENOMIC STRUCTURE

We have developed an efficient algorithm implemented in a program for a multiprocessor computing system, which makes it possible to discover genes lost or acquired during the evolution of most species from a given set. The new approach takes into account the mutual arrangement of genes on a chromosome and allows us to simultaneously consider hundreds of species. The research was carried out using supercomputers at the Joint Supercomputer Center of the Russian Academy of Sciences (JSCC RAS).

PLANNING OF DOWNLOAD PROCESSORS IN MULTIPROCESSOR SYSTEMS OF CRITICAL PURPOSE

At present, multiprocessor systems of a critical nature are widely used. Such systems are used for tracking, aiming, observing, etc. Such tasks, as a rule, require maximizing productivity and reducing the time to solve a problem. For these purposes, the initial selection of non-dependent linear, conditional and cyclic sections of sequential programs is used [1]. This is done to release fragments of programs that can be assigned to execution on processors in such a way that during execution they exchange data with neighboring processors as little as possible. Due to this, it is possible to partially improve the performance of a multiprocessor computing system, together with a decrease in the overall execution time of the entire task as a whole. For systems of the considered nature of the processor of the entire system, it is desirable to assign program fragments so that they are constantly loaded throughout the solution of the entire problem. This is another way to improve the performance of a multiprocessor system. It is obvious that the use of software for this purpose is not real due to the criticality of the time parameter. Therefore, it is relevant to use methods and corresponding hardwareoriented algorithms for scheduling processor loads, which is the subject of research in this article. The article shows the relevance of the constant loading of processors of multiprocessor systems with a high availability factor. The necessity of drawing up a plan for loading processors to support this coefficient is substantiated. An appropriate method and algorithm for multiprocessor systems for critical purposes (tracking systems, surveillance, aiming, etc.) are proposed.

Realization Features of System Software of Multiprocessor Computing Systems

The chapter is aimed at the problem of use and adjustment of system software of multiprocessor computing systems. The main principles of the Linux operating system were considered, which were necessary when constructing a multiprocessor computing system. These studies also cover new ways to remotely access the memory of processor systems through the use of RDMA technology for InfiniBand technology. Thus, it has been shown that the RDMA principle, together with the formation of a separate computing network in the data interchange environment, and the implementation of VLAN mechanisms, allowed the data transmission among nodes memory of the multiprocessor computing system without additional buffering. This approach does not require the active OS operation, libraries, or applications on those nodes of the system which memory is requested.