Hybrid MPI-Thread Parallelization of the Fast Multipole Method

Coulaud, Olivier; Fortin, Pierre; Roman, Jean

doi:10.1109/ispdc.2007.29

Cited by 8 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…9 that the parallel efficiency of FMB is superior to that of GADGET-2. Furthermore, it usually tends to increase as the number of particles increase (not shown here), since as shown in Coulaud et al (2007) the hybrid MPI-thread parallelization of FMB makes it possible to gain parallel efficiency over a pure MPI code such as GADGET-2. Figure 10 indicates the memory consumptions of the various particle distributions.…”

Section: Parallel Execution Comparisonmentioning

confidence: 88%

“…The FMB code (Fast Multipole with BLAS 2 ) is an implementation of the traditional fast multipole method (Coulaud et al 2007(Coulaud et al , 2008(Coulaud et al , 2010. This code is based on the algorithm of Nabors et al (1994): the octree height H is optimally set by the user to minimize the computation time.…”

Section: The Different Algorithmsmentioning

confidence: 99%

“…GADGET-2 relies indeed on the MPI standard for message-passing communications 3 , and FMB is a hybrid MPI-thread code (Coulaud et al 2007) that exploits both multi-process and multi-thread modes.…”

Section: The Different Algorithmsmentioning

confidence: 99%

“…Since the GADGET-2 code is a pure MPI code, we use one MPI process per core, namely up to 128 MPI processes, and one node for up to 16 MPI processes since intra-node MPI communications are faster than inter-node ones. In contrast, FMB is a hybrid MPI-thread code that is most efficient with one MPI process per node and 16 computation threads (one for each core available on the node) inside each MPI process: we refer to Coulaud et al (2007) for more details. We therefore use up to 8 MPI processes for FMB (with a total of 8 × 16 = 128 threads).…”

Section: Parallel Execution Comparisonmentioning

confidence: 99%

“…As discussed e.g. by Springel (2005) or Coulaud et al (2007), space filling curves such as the Morton or Hilbert orderings are used in practice with estimated costs (for example, roughly equal number of particles per process) or cost functions (for example, number of particles in the cell). These costs based on the particle number give only an approximation of the real computation cost thereby attributed to each process.…”

Section: Parallel Execution Comparisonmentioning

confidence: 99%

See 4 more Smart Citations

Comparisons of different codes for galacticN-body simulations

Fortin

Athanassoula²,

Lambert³

2011

A&A

View full text Add to dashboard Cite

Context. N-body simulations are widely used for galactic dynamics studies. Several different algorithms have been developed and a number of authors have made their codes public. Aims. We wish to help potential users of these codes, particularly less experienced ones, to select the appropriate code by giving them relevant information. Methods. We consider different implementations of three different algorithms, namely the Barnes-Hut tree-code algorithm (implemented both in GADGET-2 and as a NEMO program), the fast multipole method (implemented in the FMB code -Fast Multipole with BLAS) and W. Dehnen's algorithm (implemented in falcON -force algorithm with complexity O(N)). We compare extensively the timing and memory comparisons of these different implementations and discuss other advantages and disadvantages of these codes.Results. In terms of serial executions, falcON is clearly the fastest code, in many cases by a wide margin. In addition, it is much less sensitive to concentration than other codes, it conserves linear momentum, its computational costs depends roughly linearly on the number of particles, and it allows for manipulators. GADGET-2 is the second fastest and, being parallel, can outperform falcON if a sufficient number of processors is used, the number of which, in our tests is between three and ten. Its public version includes an SPH representation of the gas component. FMB is faster than the implementation of the Barnes-Hut algorithm in NEMO, as long as the particle distribution is not too concentrated, but remains considerably slower than falcON and GADGET-2. It has, nevertheless, the highest parallel efficiency and memory scalability. Conclusions. We recommend falcON for pure N-body simulations with a relatively restricted number of particles (up to a few million). In cases of a larger number of particles, a number of processors being available, or studies including gas and its physics, we recommend GADGET-2. FMB provides the highest parallel efficiency and memory scalability, but is considerably slower in execution time and is ill-adapted to the concentrated distributions encountered in galactic work.

show abstract

Section: Parallel Execution Comparisonmentioning

confidence: 88%