Development and application of distributed computing system in inverse problems of fracture mechanics

Introduction. When solving problems that require processing of a large amount of data, the problem of obtaining a solution in an acceptable time arises. One of the ways to implement a time-resource constraint is to split the entire amount of computing between several data centers. Therefore, to develop methods to improve the speed of computing systems and the effectiveness of their use is an important task. The paper deals with the development of a distributed computing system to provide solutions to inverse problems in the field of fracture mechanics. The work objectives are the design, and calculation and experimental justification of the system for solving an ad-hoc type of tasks. Such software packages as ANSYS, COMSOL, and FlexPDE are used in the work. Materials and Methods. A methodology is proposed that allows for the use of ordinary PCs as processing centers, rather than specialized machines with preinstalled hardware. The system does not impose any special requirements on the hardware. For the system operation, the communication between PCs is necessary. The availability of the high-quality high-speed network is desirable. This simplifies the process of the system deployment, and increases the productivity of the computing process. A finite element model with a large set of parameters is calculated. Research Results. New software is developed to provide solutions to inverse problems in the field of fracture mechanics. The opportunity to use the system for solving a wider range of tasks is realized. Special features of similar software are taken into account in order to increase resiliency and reduce costs not related to the problem solution. The system executing module performs calculations in a multithreaded mode. Therefore, the hardware capabilities of computing tools are used with maximum efficiency. Simplicity in organizing the format of data storage and transferring them over the network made it possible to achieve the most optimal utilization of the available resources. Discussion and Conclusions. The developed system of distributed computations is used in performing finite element modeling of the ultrasonic wave propagation in the wall of a tube with a thin outer coating. The system provides a backup of data. This made it possible to reduce the possible losses of design data when some components of the system fail.

распределенные вычисления, гетерогенные системы, конечно-элементное моделирование, неразрушающий контроль, идентификация дефектов, distributed computing, heterogeneous systems, finite element modeling, non-destructive testing, defect identification

1. Chih-Fong Tsai, Wei-Chao Lin, Shih-Wen Ke. Big data mining with parallel computing: A comparison of distributed and MapReduce methodologies. Journal of Systems and Software, 2016, vol. 122, pp. 83–92.

2. Gantz, J. The digital universe in 2020: Big data, bigger digital shadows, and biggest growth in the Far East. IDC iView: IDC Analyze the Future, 2012, vol. 2007, pp. 1–16.

3. Shvachko, K., Kuang, H., Sanjay, R., Chansler, R. The Hadoop distributed file system. Mass Storage Systems and Technologies (MSST), 2010 IEEE 26th Symposium on, IEEE, 2010, pp. 1–10.

4. Lakshman, A., Malik, P. Cassandra – A decentralized structured storage system. ACM SIGOPS Oper. Syst. Rev., 2010, vol. 44, no. 2, pp. 35–40.

5. Thusoo, A., Sarma, J.S., Jain, N., Shao, Z., Chakka, P., Anthony, S., Liu, H., Wyckoff, P., Murthy, R. Hive—A warehousing solution over a Map-Reduce framework. Proc. VLDB Endowment, 2009, vol. 2 no. 2, pp. 1626–1629.

6. Foster, I., Kesselman, C., Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations. International Journal of Supercomputer Applications and High-Performance Computing, 2001, vol. 15, no 3, pp. 200–222.

7. Seinstra, F., Maassen, J., van Nieuwpoort, R.V., Drost, N., van Kessel, T., van Werkhoven, B., Urbani, J., Jacobs, C., Kielmann, T., Bal, H.E. Jungle computing: Distributed supercomputing beyond clusters, grids, and clouds. Grids, Clouds and Virtualization, London: Springer, 2011, pp. 167–197.

8. Tanenbaum, E. Raspredelennye sistemy: printsipy i paradigmy. [Distributed systems: principles and paradigms.] St.Petersburg: Piter, 2003, 877 p. (in Russian).

9. Soloviev, A.N., Sobol, B.V., Vasiliev, P.V. Ultrasonic Location of Inner Crack Defects in a Compound Elastic Cylinder Using an Artificial Neural-Network Apparatus. Russian Journal of Nondestructive Testing, 2016, vol. 52, no. 3, pp. 119–124.

10. Vatulyan, А.О. Obratnye zadachi v mekhanike deformiruemogo tverdogo tela. [Inverse problems in deformable solid mechanics.] Moscow: Fizmatlit, 2007, 224 p. (in Russian).

11. Willcox, M.A. A Brief Description of NDT Techniques. Toronto: NDT Equipment Limited, 2003, 54 p.