Implementation of parallel I/O in DVM-system
V.A. Bakhtin, V.A. Krukov, N.V. Podderugina, M.N. Pritula
DVM-system was designed to create parallel programs of scientific-technical calculations in C-DVMH and Fortran-DVMH languages. These languages use the same model of parallel programming (DVMH-model) and are the extensions of standard C and Fortran languages by parallelism specifications, implemented as compiler directives. DVMH-model allows to create efficient parallel programs for heterogeneous computational clusters, which nodes use as computing devices not only universal multi-core processors but also can use attached accelerators (GPUs or Intel Xeon Phi coprocessors). This article discusses new possibilities of parallel I/O, which were implemented in the DVM-system recently. Using the developed I/O subsystem can significantly speed up DVMH-applications by combining input-output operations with computations.
аutomation of parallel program development, DVM-system, parallel I/O, checkpoint, Фортран, C.
V.A. Bakhtin, V.A. Krukov, N.V. Podderugina, M.N. Pritula. Implementation of parallel I/O in DVM-system // Russian Supercomputing Days: Proccedings of International conference (26-27 september 2016, Moscow), Moscow: Moscow University Press, 2016, P. 615-625