Parallel and distributed computation: numerical methods
Parallel and distributed computation: numerical methods
An asynchronous parallel Newton method
Mathematical Programming: Series A and B
A new method for solving triangular systems on distributed-memory message-passing multiprocessors
SIAM Journal on Scientific and Statistical Computing
Direct methods for sparse matrices
Direct methods for sparse matrices
Parallel block-partitioning of truncated Newton for nonlinear network optimization
SIAM Journal on Scientific and Statistical Computing
Parallel Block Triangular Decompositions for Solving Sparse Nonlinear Systems of Equations
Proceedings of the Fifth SIAM Conference on Parallel Processing for Scientific Computing
Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)
| Hi-index | 0.00 |
One important process in oil refining is to separate the crude oil into various oil products. This process is called distillation. In designing a complex distillation column, a large computer simulation is conducted. This paper presents our experience with parallelizing an oil refining simulation application that computes the composition of the various oil products in designed refining columns operated under a given set of conditions. Mathematical models for the simulation form large sparse nonlinear systems of equations. Triangular decompositions of sparse nonlinear systems are fundamental numerical methods in this simulation computing. Different approaches have been applied to carry out this simulation in parallel. Parallelisms of the simulation are exploited at three levels—direct parallelization, structured parallelization and a synchronous parallelization of the problem. The approaches of this practical research provide insight into some important issues of parallel computing for real-world applications. The parallel programs were implemented and run on the Intel iPSC/860. Parallel computing results are presented with comparisons and discussions.