Master DonNTU Sereda Oleg

Abstract

Сontent

Introduction
1. Theme urgency
2. Goal and tasks of the reseach
3. The research of model of a dybanical system on a multi-processor platform
Conclusion
References

Introduction

 The idea of parallel processing as a powerful reserve increased performance of computing devices was proposed by Charles Babbage in about one hundred years before the advent of the first electronic computer. However, the level of technology the mid-19th century did not allow him to realize this idea. With the advent of the first electronic computers, these ideas often became the starting point for the development of the most advanced and productive computing. No exaggeration to say that the whole history of the development of high-performance computing systems — a history of the ideas of parallel processing at some stage in the development of computer technology, of course, combined with an increase in the frequency of electronic circuits.

1. Theme urgency

 Development of software and hardware for research is more complex dynamic systems model provides support for object technology in the field of mechanics of solids, molecular dynamics, chemical industry, electronics, and others. In the time domain simulation systems and objects usually describes systems of differential and - differential algebraic equations. For the numerical integration of stiff and semi-natural objects in real time requires considerable computing power. One way to improve the performance of modeling systems use multi-processor, in particular cluster(Massive Parallel Processing), systems.

 A characteristic feature of MPP multiprocessor architecture is that memory is divided and each node has access only to his memory. Like any architecture MPP has problem. The first problem is the lack of shared memory reduces the speed of the network. The second problem is that each node uses a limited amount of memory. The third problem is the complexity of the maximum resources.

 Master's work is devoted to the actual scientific problem — improving the efficiency and the expansion of the class of tasks through the development of special Scheduler software and monitor system performance for multi-processor MPP system. Investigation of complex dynamic systems based on multi-processor computer will analyze the existing conflicts with using of memory and develop parallel techniques of data flows and optimize performance in the complex dynamic system on multi-processor MPP-platform.

2. Goal and tasks of the reseach

 The aim of this study is to develop a special software scheduler and monitor implementation of the system for multi-processor MPP computers.
 The main tasks of the researching:

1. An analytical model describing the analysis of dynamic systems.
2. The development of the block diagram of parallel model of a dynamic system and improving the exchange of information in this model.
3. The implementation of a multi-frequency monitor control dynamical system.
4. The research of the temporal characteristics of computing and exchanges with standard programming tools.
5. Development of the scheduler and monitor multi-frequency dynamic control system.

 The object of research: complex dynamic system for multi-processor MPP-platform.
 The subject of research: development of models and special scheduler software , expanding the class of tasks.
 Scientific novelty consists in develop own software for parallel tasks on the cluster system.
 As part of the master's work is planned to obtain relevant scientific results in the following areas:

1. Improving the efficiency of information exchange in the parallel model of a dynamic system.
2. The parallel technique of data flows with the small volumes.
3. Expanding the class of tasks on MPP-platform and method of creating programs for this class.

3. The research of model of a dybanical system on a multi-processor platform

 A computer cluster consists of a set of loosely connected or tightly connected computers that work together so that in many respects they can be viewed as a single system.

 The components of a cluster are usually connected to each other through fast local area networks (“LAN”), with each node (computer used as a server) running its own instance of an operating system. Computer clusters emerged as a result of convergence of a number of computing trends including the availability of low cost microprocessors, high speed networks, and software for high performance distributed computing.   Clusters are usually deployed to improve performance and availability over that of a single computer, while typically being much more cost-effective than single computers of comparable speed or availability [1].

A cluster of Donetsk National Technical University is a computer system MPP - architecture. Architecture cluster DonNTU is shown in Figure 1.

Figure 1 – Architecture cluster DonNTU.

 On a cluster of Donetsk National Technical University used OS ScientificLinux 5.4. To transfer and synchronize exchange using UDP and broadcast, and as a means of transmission - sockets.
 To create a software model to performed numerical integration of the system of differential equations obtained by reduction of the order of equations Vanderpolya [2]:

 As you can see, the above system is of second order. There are some basic techniques of numerical solutions of systems of differential equations. We have chosen numerical method Euler formula trapezoids, because of the simplicity of the implementation in an iterative cycle. For us, there is not a critical error in the method. The essence of the method is outlined in Figure 2.

Figure 2 – Numerical method Euler formula trapezoids.

 To create a dynamic object model uses a system of Vanderpol equations of the first order, obtained after the application of the Euler method:

 The result was a system of parallel processes on the cluster. The system and the results can be seen in Figures 3 and 4.

Figure 3 – Develop parallel processes system.


Figure 4 – Results of system work.

Conclusion

 Studies have shown performance software developed and demonstrated the possibility of simultaneous solutions of several equations. The correctness of the chosen solutions of the dynamical system can be seen on the following figures.

References

 1. Materials from the free encyclopedia “Wikipedia” — Computer cluster [electronic resource]. – Access mode: http://en.wikipedia.org/wiki/Computer_cluster.

 2. Максименко В. Л. Исследование и разработка специальных программных способов для моделирования автоколебаний. Портал магистров ДонНТУ.

 3. Букатов А.А., Дацюк В.Н., Жегуло А.И. “Программирование многопроцессорных вычислительных систем”, Ростов-на-Дону, 2003 г.

 4. Материалы из свободной энциклопедии “Википедия”; — Гармонический осциллятор [Электронный ресурс]. – Рeжим дoступa к ресурсу: http://ru.wikipedia.org/wiki/Гармонический_осциллятор.

 5. Магистерская работа Ю.А. Иванова“Исследование и разработка алгоритмов планирования процессов реального времени моделирования сложных динамических систем”.

 6. Параллельная вычислительная система – кластер NeClus[электронный ресурс]. Тип доступа: http://neclus.donntu.ru/

 7. Шон Уолтон Создание сетевых приложений в среде Linux. Руководство разработчика: Москва/Санкт-Питербург/Киев, 2001г. – 116-213 с., 358-379с.