부산대학교 도서관

In this work, the authors investigate the scalability of a MPI approach for solving a nonlinear system of ODEs by Euler's method. The system describes SEIRS (Susceptible-Exposed-Infectious-Recovered-Susceptible) epidemic disease in the prey where the predator-prey interaction is given by Lotka - Volterra type. The different parameters of the system are grouped in four groups: (i) migration parameters; (ii) natural mortality parameters; (iii) mortality by disease and predation and (iv) specific parameters describing infection rate, incubation rate, recovery rate, etc. All parameters grouping in the above 4 groups are discretized with a fix step in a given interval. Thus, a huge number of different combinations of the parameters' values of the system is received corresponding to given equilibrium points. The parallel algorithm allow us to receive a large number of solutions of the system of ODEs. Using these solutions we can select those cases of system's parameters in which the dynamics of the population is stable and the disease is controlled. A lot of parallel tests were done on three HPC systems: ARGO HPC system in the Department of Informatics and Telecommunications of the University of Athens (DIT-UA, Greece), ARIS HPC system in the National Infrastructures for Research and Technology S.A. (GRNET S.A., Greece), and Avitohol supercomputer in the Institute of Information and Communication Technologies (IICT, Bulgaria). The numerical results show that the scalability depends on the architecture of the HPC system and on I/O file system usage from the parallel algorithm. [ABSTRACT FROM AUTHOR]