Massively parallelized interpolated factored Green function method
نویسندگان
چکیده
This paper presents the first parallel implementation of novel “Interpolated Factored Green Function” (IFGF) method introduced recently for accelerated evaluation discrete integral operators arising in wave scattering and other areas (Bauinger Bruno, Jour. Computat. Phys., 2021). On basis hierarchical IFGF interpolation strategy, proposed (hybrid MPI-OpenMP) results efficient data communication, it scales up to large numbers cores—without any hard limitations on number cores efficiently employed. Moreover, given cores, approach preserves O(NlogN) computing cost inherent sequential version algorithm. Unlike approaches, does not utilize Fast Fourier Transform (FFT), is thus better suited parallelization distributed-memory computer systems. In particular, relies a “peer-to-peer” strategy wherein, at every level, field propagation directly enacted via “exchanges” between “peer” polynomials constant degree, without accumulation large-scale “telephone-central” mathematical constructs such as those used Multipole Method (FMM) pure FFT-based approaches. A variety numerical presented this illustrate character algorithm, particular demonstrating scaling from 1 all 1,680 available High Performance Computing cluster used, problems 4,096 wavelengths acoustic size.
منابع مشابه
Interpolated Backoff for Factored Translation Models
We propose interpolated backoff methods to strike the balance between traditional surface form translation models and factored models that decompose translation into lemma and morphological feature mapping steps. We show that this approach improves translation quality by 0.5 BLEU (German–English) over phrase-based models, due to the better translation of rare nouns and adjectives.
متن کاملThe parallelized Pollard kangaroo method in real quadratic function fields
We show how to use the parallelized kangaroo method for computing invariants in real quadratic function fields. Specifically, we show how to apply the kangaroo method to the infrastructure in these fields. We also show how to speed up the computation by using heuristics on the distribution of the divisor class number, and by using the relatively inexpensive baby steps in the real quadratic mode...
متن کاملMassively parallelized replica-exchange simulations of polymers on GPUs
We discuss the advantages of parallelization by multithreading on graphics processing units (GPUs) for parallel tempering Monte Carlo computer simulations of an exemplified bead-spring model for homopolymers. Since the sampling of a large ensemble of conformations is a prerequisite for the precise estimation of statistical quantities such as typical indicators for conformational transitions lik...
متن کاملelectronic transmission wave function of disordered graphene by direct method and green's function method
we describe how to obtain electronic transport properties of disordered graphene, including the tight binding model and nearest neighbor hopping. we present a new method for computing, electronic transport wave function and greens function of the disordered graphene. in this method, based on the small rectangular approximation, break up the potential barriers in to small parts. then using the f...
متن کاملScalability of the Parallelized Pollard Rho Method
The integrity of elliptic curve cryptography (ECC) stems from the presumed complexity of the elliptic curve discrete logarithm problem (ECDLP). Thus, the study of algorithms to solve the ECDLP is essential for two reasons: (1) to establish the integrity of ECC, and (2) to guide designers in the choice of ECC parameters, such as the length of the parameters and the characteristics of the curve. ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Journal of Computational Physics
سال: 2023
ISSN: ['1090-2716', '0021-9991']
DOI: https://doi.org/10.1016/j.jcp.2022.111837