An OpenMP parallel implementation using a coprocessor for numerical simulation of oil reservoirs

Abstract

The objective of this work is to parallelize, using the Application Programming Interface (API) OpenMP (Open Multi-Processing) and Intel Xeon Phi coprocessor based on Intel Many Integrated Core (MIC) architecture, the numerical method used to solve the algebraic system resulting from the discretization of the differential partial equation (PDE) that describes the single-phase flow in an oil reservoir. The set of governing equations are the continuity equation and the Darcy's law. The Hydraulic Diffusivity Equation (HDE), for the unknown pressure, is obtained from these fundamental equations and it is discretized by means of the Finite Difference Method (FDM) along with a time implicit formulation. Different numerical tests were performed to study the computational efficiency of the parallelized versions of Conjugate Gradient, BiConjugate Gradient and BiConjugate Gradient Stabilized methods. Speedup results were considered to evaluate the performance of the parallel algorithms for the horizontal well simulation case. The methodology also included a sensibility analysis for different production scenarios including variations on the permeability, formation-valuefactor, well length and production rate.

Parallel Multi-Stage Preconditioners with Adaptive Setup for the Black Oil Model

“…As multi-core and many-core architectures become more and more popular, parallel computing for petroleum reservoir simulation is now a subject of great interest. In recent years, there has been some work on parallel algorithms for reservoir problems; see [33][34][35][36][37][38][39][40][41][42][43][44][45] and references therein. For example, Feng et al [35] designed an OpenMP parallel algorithm with high efficiency and low memory cost for standard interpolation and coarse grid operator of AMG, under the framework of Fast Auxiliary Space Preconditioning (FASP) [46].…”

Multithreaded Acceleration of 3D Mathematical Model for Ore Sintering

“…The authors of the paper [5] successfully parallelized different variants of iterative Conjugate Gradient methods for solving finite-difference implicit problem using an OpenMP API with 228 threads and Intel Xeon Phi with 57 cores. A good refining of a numeric grid significantly increases effectiveness of the OpenMP implementation.…”

Uma implementação paralelizada via a API OpenMP para a simulação numérica de reservatórios de gás natural

“…Diversas áreas de pesquisa e desenvolvimento, além de aplicações industriais, têm tirado proveito da aplicação de técnicas tais como o OpenMP. Por exemplo, a simulação numérica de reservatórios de óleo e/ou gás (Werneck, 2016, Redondo, 2017, Werneck et al, 2019, sendo que os reservatórios de gás natural são o objeto de interesse deste artigo.…”

“…Ressalta-se que existem métodos mais robustos, que são empregados correntemente na simulação de reservatórios em escala de campo, como os nãoestacionários (Ertekin et al, 2001). Os leitores podem se reportar ao artigo de Werneck et al (2019) para mais informações sobre a paralelização, usando o OpenMP e um coprocessador, dos métodos não-estacionários dos Gradientes Conjugados, Gradiente Biconjugado e Gradiente Biconjugado Estabilizado.…”

“…É interessante notar que para os testes variando a vazão e a densidade há a ocorrência de tempos de execução maiores para o uso de 4 threads do que para duas, o que certamente pode ser explicado em função das características intrínsecas ao modelo fork-join e às cláusulas de barreira (Werneck et al, 2019).…”