Parallelization of ITOUGH2 using PVM

نویسنده

  • Stefan Finsterle
چکیده

ITOUGH2 inversions are computationally intensive because the forward problem must be solved many times to evaluate the objective function for different parameter combinations or to numerically calculate sensitivity coefficients. Most of these forward runs are independent from each other and can therefore be performed in parallel. Message passing based on the Parallel Virtual Machine (PVM) system has been implemented into ROUGH2 to enable parallel processing of ITOUGH2 jobs on a heterogeneous network of Unix workstations. This report describes the PVM system and its implementation into ITOUGH2. Instructions are given for installing PVM, compiling ITOUGH2-PVM for use on a workstation cluster, the preparation of an ITOUGH2 input file under PVM, and the execution of an ITOUGHZPVM application. Examples are discussed, demonstrating the use of ITOUGH2-PVM. ITOUGH2-PVM i ABSTRACT

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Automatic Calibration of Geothermal Reservoir Models through Parallel Computing on a Workstation Cluster

ITOUGH2 is an optimization code that allows estimation of any input parameter of the nonisothermal, multiphase flow simulator TOUGH2. ITOUGH2 inversions are computationally intensive because the so-called forward problem, i.e., the simulation of fluid and heat flow through the geologic formation, must be solved many times for different parameter combinations to evaluate the misfit criterion or ...

متن کامل

JPT: A Java Parallelization Tool

PVM is a succesfull programming environment for distribu ted computing in the languages C and Fortran Recently several imple mentations of PVM for Java have been added making PVM program ming accessible to the Java community With PVM for Java however the user still needs to partition the prob lem calculate the data partitioning and program the message passing and synchronization In this paper J...

متن کامل

PVM–based Training of Large Neural Architectures

In this paper a methodology for parallelizing neural network training algorithms is described, based on the parallel evaluation of the error function and gradient using the Parallel Virtual Machine (PVM). PVM is an integrated set of software tools and libraries that emulates a general–purpose, flexible, heterogeneous concurrent computing framework on interconnected computers of varied architect...

متن کامل

Parallel Acceleration and Rebalancing Schemes for Solving Transport Problems Using Pvm

| In this paper, parallelization for CP calculation and multigroup ux computation are presented. Implementation of acceleration and neutron rebalancing strategies is also investigated. Particular techniques pertinent to the two-step energy/space iterative process of solving a multigroup transport equation are described. The parallel performance is studied in cases where the cyclic tracking tech...

متن کامل

Parallel Domain Decomposition for 1-D Active Thermal Control Problem with PVM

this paper describes a 1-D Active Thermal Control Problem (1-D ATCP) with the use of Stationary Iterative Techniques (Jacobi and Gauss-Seidel) on the discretization of the resulted matrices. Parallelization of the problem is carried out using Domain Decomposition (DD) parallel communication approach with Parallel Virtual Machine (PVM) to enable better flexibility in parallel execution, and grea...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1998