Fortran 2008 coarrays

Abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms Fortran 2008 coarrays ABSTRACTCoarrays are a Fortran 2008 standard feature intended for SPMD type parallel programming. The runtime environment starts a number of identical executable images of the coarray program, on multiple processors, which could be actual physical processors or threads. Each image has a unique number and its private address space. Ordinary variables are private to an image. Coarray variables are available for read/write access from anyo ther image. Coarray communications are of "single sided" type, i.e. a remote call from image A to image B does not need to be accompanied by a corresponding call in image B. This feature makes coarray programming a lot simpler than MPI. The standard provides synchronisation intrinsics to help avoid race conditions or deadlocks. Any ordinary variable can be made into a coarray -scalars, arrays, intrinsic or derivedd ata types, pointers, allocatables are all allowed. Coarrays can be declared in, and passed to, procedures. Coarrays are thus very flexible and can be used for a number of purposes. Forexample a collection of coarrays from all or some images can be thought of as a large single array.T his is precisely the inverse of the model partitioning logic, typical in MPI programs. A coarray program can exploit functional parallelism too, by delegating distinct tasks to separate images or teams of images. Coarray collectivesa re expected to become a part of the next version of the Fortran standard. Am ajor unresolved problem of coarray programming is the lack of standard parallel I/O facility in Fortran. In this paper several simple complete coarray programs are shown and compared to alternative parallel technologies -OpenMP,M PI and Fortran 2008 intrinsic "do concurrent". Inter image communication patterns and data transfer are illustrated. An example of a materials microstructure simulation coarray program scaled up to 32k cores is shown. Problems with coarray I/O at this scale are highlighted and addressed with the use of MPI-I/O. A hybrid MPI/coarray programming is discussed and illustrated with a finite element/cellular automata (CAFE) multi-scale model. The paper completes with a description of the newc oarray language features, expected in the 2015 Fortran standard, and with a brief list of coarray resources.7April 2015

Towards mechanism-based simulation of impact damage using exascale computing

“…In this work we use ParaFEM FE library, parafem.org.uk [12] and CGPACK CA library, cgpack.sourceforge.net [8,13]. Both libraries scale well on massively parallel computer, see Figure 1.…”

“…There is an emerging opinion in HPC circles that PGAS languages and one-way communication patterns are better suited for exascale systems than MPI [15,16]. A bi-directional FE to CA interface layer was created which maps MPI ranks to coarray images, based on the concept that each element of volume is represented by both the FE and the CA [13]. At present the scaling of the CAFE model is worse than when each component library is used independently, see Figure 1.…”

Determination of basic calculation and experimental parameters of device for bore hole cleaning

“…Уравнение движения жидкости в абсолютной си-стеме координат для любой произвольной точки примет вид (Shterenlikht, 2008):…”

“…Экспериментальные значения осевой составляющей скорости потока заметно превышают таковые, полученные аналитическим путем, причем эта тенденция справедлива для всего диапазона ра-циональных величин угловых характеристик лопаст-ного элемента (Ighnatov, 2016). За их пределами кар-тина течения активного потока резко изменяется, в особенности для ньютоновских жидкостей (Shterenlikht, 2008). Наличие определенных структурно-механических свойств положительно сказывается на процессе формирования струй, причем, в отдельных случаях, даже для тех форм лопастного органа, для которых при использовании воды устойчивый режим течения не наблюдается.…”

Multi-scale CAFE Framework for Simulating Fracture in Heterogeneous Materials Implemented in Fortran Co-arrays and MPI

“…A finite CA space is most useful if used with regular boundaries, typically periodic (self-similar) or fixed. Coarrays, a Fortran native means for SPMD programming [13]- [15], are a natural implementation choice for CA models. A 3D CA space with cubic cells of discrete states maps perfectly onto a 3D integer array coarray.…”