Despite fast development of computer technology some numerical problems are still too complex to be solved using personal computers. In such cases the use of supercomputers can be a good choice. However, running jobs at computer centers involves necessity of queuing. Additionally, problems with fast transmission may arise if output files are large enough. An alternative solution is to build multiprocessor clusters of computers on which Message Passing Interface is implemented. In this paper application of MPI is presented for the case of the FLASH code which solves initial-value problems for magnetohydrodynamic equations
International audienceThe Fast Multipole Method (FMM) is considered as one of the top ten algorithms...
Abstract—Data movement in high-performance computing systems accelerated by graphics processing unit...
Computer simulations of N-body systems are beneficial to study the overall behavior of a number of p...
The universe has always been object of our fascination. Because the Sun is our nearest star naturall...
We use a simple message-passing (MPI) algorithm on a parallel cluster to reduce the simulation times...
This paper introduces the Sheffield Magnetohydrodynamics Algorithm Using GPUs (SMAUG+), an advanced ...
We present a tree-code for integrating the equations of the motion of collisionless systems, which h...
We describe a methodology for developing high performance programs running on clusters of SMP no...
Workstation clusters have become an increasingly popular alternative to traditional parallel superco...
Unconventional multi-core processors (e.g., IBM Cell B/E and NYIDIDA GPU) have emerged as accelerato...
Several clustering methods based on optimisation have been developed recently. One of them is based ...
AbstractThis paper presents a dynamic domain decomposition (D3) technique for implementing the paral...
In this work, we propose a method that allows us to decrease the energy consumption in supercomputin...
The various approaches to improving the speeds of computers by exploiting parallelism is briefly de...
The mixing of shared memory and message passing programming models within a single application has o...
International audienceThe Fast Multipole Method (FMM) is considered as one of the top ten algorithms...
Abstract—Data movement in high-performance computing systems accelerated by graphics processing unit...
Computer simulations of N-body systems are beneficial to study the overall behavior of a number of p...
The universe has always been object of our fascination. Because the Sun is our nearest star naturall...
We use a simple message-passing (MPI) algorithm on a parallel cluster to reduce the simulation times...
This paper introduces the Sheffield Magnetohydrodynamics Algorithm Using GPUs (SMAUG+), an advanced ...
We present a tree-code for integrating the equations of the motion of collisionless systems, which h...
We describe a methodology for developing high performance programs running on clusters of SMP no...
Workstation clusters have become an increasingly popular alternative to traditional parallel superco...
Unconventional multi-core processors (e.g., IBM Cell B/E and NYIDIDA GPU) have emerged as accelerato...
Several clustering methods based on optimisation have been developed recently. One of them is based ...
AbstractThis paper presents a dynamic domain decomposition (D3) technique for implementing the paral...
In this work, we propose a method that allows us to decrease the energy consumption in supercomputin...
The various approaches to improving the speeds of computers by exploiting parallelism is briefly de...
The mixing of shared memory and message passing programming models within a single application has o...
International audienceThe Fast Multipole Method (FMM) is considered as one of the top ten algorithms...
Abstract—Data movement in high-performance computing systems accelerated by graphics processing unit...
Computer simulations of N-body systems are beneficial to study the overall behavior of a number of p...