From HP-SEE Wiki

Revision as of 14:57, 14 March 2012 (view source) Mid (Talk | contribs) ← Older edit		Revision as of 10:01, 22 March 2012 (view source) Lifesci (Talk | contribs) (→OpenMPI) Newer edit →
Line 16:		Line 16:
	== OpenMPI ==		== OpenMPI ==
-		+
		+
		+	CMSLTM @IMBB (NEURON ParallelContext)
		+	Preferred MPI environment is OpenMPI (openmpi_gcc-1.4.3). NEURON was compiled with parallel support (MPI) using the gcc compiler:
		+
		+	./configure --without-iv --with-paranrn --prefix=/home/gkastel/src/nrn-7.1
		+
		+	We used NEURON's ParallelContext to distribute the simulation of each neuron to different nodes evenly.
		+
	== Porting between MPI implementations ==		== Porting between MPI implementations ==

---

Revision as of 10:01, 22 March 2012

Contents 1 Message passing model 1.1 MPICH implementations 1.1.1 MPICH 1.1.2 MVAPICH 1.1.3 MPICH-MX 1.1.4 MPICH2 1.1.5 Intel MPI 1.1.6 MPIX (IBM) 1.2 OpenMPI 1.3 Porting between MPI implementations 2 Shared memory model 2.1 OpenMP 2.2 Pthreads 3 Hybrid programming 4 CUDA

Message passing model

MPICH implementations

MPICH

MVAPICH

MPICH-MX

MPICH2

Intel MPI

MPIX (IBM)

OpenMPI

CMSLTM @IMBB (NEURON ParallelContext) Preferred MPI environment is OpenMPI (openmpi_gcc-1.4.3). NEURON was compiled with parallel support (MPI) using the gcc compiler:

 ./configure --without-iv --with-paranrn --prefix=/home/gkastel/src/nrn-7.1

We used NEURON's ParallelContext to distribute the simulation of each neuron to different nodes evenly.

Porting between MPI implementations

Shared memory model

OpenMP

Pthreads

Hybrid programming

CUDA