This paper will describe the application of the PGAS Global Arrays (GA) library to power grid simulations. The GridPACK (TM) framework has been designed to enable power grid engineers to develop parallelsimulations o...
详细信息
ISBN:
(纸本)9781509052141
This paper will describe the application of the PGAS Global Arrays (GA) library to power grid simulations. The GridPACK (TM) framework has been designed to enable power grid engineers to develop parallelsimulations of the power grid by providing a set of templates and libraries that encapsulate most of the details of parallel programming in higher level abstractions. The communication portions of the framework are implemented using a combination of message-passing (MPI) and one-sided communication (GA). This paper will provide a brief overview of GA and describe in detail the implementation of collective hash tables, which are used in many power grid applications to match data with a previously distributed network.
The proceedings contain 67 papers. The topics discussed include: studying the effectiveness of graphics accelerators paralleling solutions of three-dimensional boundary value problems on quasistructured grids;numerica...
The proceedings contain 67 papers. The topics discussed include: studying the effectiveness of graphics accelerators paralleling solutions of three-dimensional boundary value problems on quasistructured grids;numerical hydrodynamics simulation of astrophysical flows at Intel xeon phi supercomputers;approximate algorithm for choosing the best subset of nodes in the Angara interconnect with failures;parallel realization of multiscale approach for calculating the gas flows in microchannels of technical systems;distributed visualization in application to the molecular dynamics simulation of equilibrium state in the gas-metal microsystems;numerical simulation of 3d unsteady free-surface flows: development and application of a specialized parallel code;spatially efficient tree layout for GPU ray-tracing of constructive solid geometry scenes;analysis of methods for solving inverse problems of chemical kinetics with the use of parallel computing;parallelization of calculation the kinetic model of selective hydrogenation of acetylene on a gold clusters;the development of the technology platform on the basis of a supercomputer to handle large flows of experimental data;and parallel algorithms for synthesizing and processing of 3d radio holographic images.
In this paper a software package for processing and visualization of the molecular dynamics simulation is presented. During the development process the performance issues of the Python programming language were studie...
详细信息
In this paper a software package for processing and visualization of the molecular dynamics simulation is presented. During the development process the performance issues of the Python programming language were studied, as well as the abilities to avoid them. It was also studied the problem of obtaining and processing distributed data stored on multiple computational nodes. An open-source package for scientific visualization Mayavi2 was chosen as a visualization tool. As a result of application this system to the data obtained from MD simulation of the gas and metal plate interaction we were able to observe in the details the effect of adsorption, which is important for many practical applications.
Rapid pace of innovation in industrial research labs requires fast algorithm evaluation cycles. The use of multi-core hardware and distributed clusters is essential to achieve reasonable turnaround times for high-load...
详细信息
Heart simulation is complex task that requires multiscale modeling on cell, tissue and organ levels. Such structure makes difficult high performance code development and its maintenance. In this paper, we evaluate how...
详细信息
Heart simulation is complex task that requires multiscale modeling on cell, tissue and organ levels. Such structure makes difficult high performance code development and its maintenance. In this paper, we evaluate how scientiffic software could be used for heart simulation. An overview of existing frameworks for automated scientiffic computing is presented. The FEniCS framework was chosen since it supports auto-mated solution of differential equations by the finite element method of parallel computing systems, provides near-mathematical notation, uses high performance backend and has comprehensive documentation. FEniCS performance was evaluated by simulation the space propaga-tion of membrane potential alternation over a cardiac left ventricle using the electrophysiological model of a left ventricle and the Ekaterinburg-Oxford cell model. The FEniCS framework showed good performance and near-linear scalability up to 240 CPU cores.
Molecular dynamics simulation should be combined with modern supercomputer technologies to obtain meaningful results in reasonable time. Widely used simulation software in MD is LAMMPS. It produces a dump with computa...
详细信息
Molecular dynamics simulation should be combined with modern supercomputer technologies to obtain meaningful results in reasonable time. Widely used simulation software in MD is LAMMPS. It produces a dump with computational results of simulation, and such a large amount of data should be post-processed. However, LAMMPS does not provide ready-to-use utilities that would do it. In this paper we present the developed algorithm for LAMMPS output data processing for nucleation process. The analysis of implementation performance shows that it is shading that takes most of the time, hence, it should provide the greatest room for future optimization.
We developed parallel algorithm for solving Vlasov-Poisson systems of equations using particle-in-cell method. It uses new technique of dynamic load balancing for processors, which are distributed between subdomains i...
详细信息
We developed parallel algorithm for solving Vlasov-Poisson systems of equations using particle-in-cell method. It uses new technique of dynamic load balancing for processors, which are distributed between subdomains in correspndance with the number of modeling particles located in the subdomain. Domain decomposition method combines grid (eulerian) method for solving Poisson equation with lagrangian paricle method for solving Vlasov equation. It takes into account physical features of the modeling nonstationary rotating disks (both 2D and 3D).
The proceedings contain 11 papers. The topics discussed include: enabling interactive scientific data visualization and analysis with see-through HMDs and a large tiled display;collaborative distributed cognition usin...
ISBN:
(纸本)9781509008346
The proceedings contain 11 papers. The topics discussed include: enabling interactive scientific data visualization and analysis with see-through HMDs and a large tiled display;collaborative distributed cognition using a seamless desktop infrastructure;immersive analytics: crossing the gulfs with high-performance visualization;simulation exploration through immersive parallel planes;towards mobile immersive analysis: a study of applications;ShodanVR: immersive visualization of text records from the shodan database;be the data: a new approach for immersive analytics;and immersive visual analytics for transformative neutron scattering science.
Large-scale simulations play a central role in science and the industry. Several challenges occur when building simulation software, because simulations require complex software developed in a dynamic construction pro...
详细信息
ISBN:
(纸本)9781509052240
Large-scale simulations play a central role in science and the industry. Several challenges occur when building simulation software, because simulations require complex software developed in a dynamic construction process. That is why simulation software engineering (SSE) is emerging lately as a research focus. The dichotomous trade-off between scalability and efficiency (SE) on the one hand and maintainability and portability (MP) on the other hand is one of the core challenges. We report on the SE/MP trade-off in the context of an ongoing systematic literature review (SLR). After characterizing the issue of the SE/MP trade-off using two examples from our own research, we (1) review the 33 identified articles that assess the trade-off, (2) summarize the proposed solutions for the trade-off, and (3) discuss the findings for SSE and future work. Overall, we see evidence for the SE/MP trade-off and first solution approaches. However, a strong empirical foundation has yet to be established;general quantitative metrics and methods supporting software developers in addressing the trade-off have to be developed. We foresee considerable future work in SSE across scientific communities.
The proceedings contain 30 papers. The special focus in this conference is on TAPEMS 2016 and BigTrust 2016. The topics include: An in-memory event tracing extension to the open trace format 2;network-aware optimizati...
ISBN:
(纸本)9783319499550
The proceedings contain 30 papers. The special focus in this conference is on TAPEMS 2016 and BigTrust 2016. The topics include: An in-memory event tracing extension to the open trace format 2;network-aware optimization of MPDATA on homogeneous multi-core clusters with heterogeneous network;formalizing data locality in task parallel applications;improving the energy efficiency of evolutionary multi-objective algorithms;a parallel model for heterogeneous cluster;comparative analysis of OpenACC compilers;traffic sign recognition based on parameter-free detector and multi-modal representation;reversible data hiding using non-local means prediction;secure data access in hadoop using elliptic curve cryptography;statistical analysis of CCM.M-K1 international comparison based on monte carlo method;redundancy elimination in the ExaStencils code generator;a dataflow IR for memory efficient RIPL compilation to FPGAs;exploring a distributed iterative reconstructor based on split bregman using PETSc;implementation of the beamformer algorithm for the NVIDIA jetson;efficiency of GPUs for relational database engine processing;geocon;a middleware for location-aware ubiquitous applications and cellular ANTomata as engines for highly parallel pattern processing.
暂无评论