A Parallel Solver for the Design of Oil Filters

Abstract

Nowadays, it is widely recognized that computer simulation plays a crucial role in designing oil filters used in the automotive industry. However, even a single direct simulation of the flow usually requires significant computational resources. Thus, it is obvious that solution of optimization problems is only feasible using parallel computers and algorithms.In this paper, we present a general master-slave parallel template, which was specially designed for the easy integration of direct parallel solvers into a parallel optimization tool. We show how an already existing direct solver for the 3D simulation of flow through the oil filter is integrated into our template to obtain a parallel optimization solver. Some capabilities and performance of this solver are demonstrated by solving geometry optimization problem of a filter element.

Predictor-Corrector Domain Decomposition Algorithm for Parabolic Problems on Graphs

“…Similar techniques also can be applied for problems on graph structures (see [9,11] for related discussions). We also mention a recent paper [12], where the domain decomposition method is used to parallelize a solver for simulation of flows in oil filters.…”

Parallel Optimization Algorithm for Competitive Facility Location

“…The proposed distributed-memory parallel algorithm is based on the master-slave model [12], where one of the processors is responsible for the organization of the workload of the slaves as well as for the communication between them. For the application of the masterslave model to parallelize genetic and other algorithms we refer to [3,7,21]. The main concern in development of the distributed-memory parallel algorithm is the organization of the effective workload of the slave (and master) processing units, what is not a trivial task due to the nature of the algorithm being parallelized.…”

Netiesinių matematinių modelių grafuose skaitinė analizė

“…Ši tendencija prieštarauja bendram išplečiamumo įverčiui, kuris teigia, kad lygiagrečiojo algoritmo efektyvumas turėtų didėti, didinant skaičiavimo uždavinį fiksuotam procesų skaičiui. Šis efektas plačiau yra nagrinėjamas darbe (Starikovičius et al 2011).…”