Optimal memory minimization algorithms for the multifrontal method
نویسنده
چکیده
We are interested in the active and total memory usage of the mul-tifrontal method. Starting from the algorithms proposed by Liu, we suggest a new scheme together with a tree traversal that give an optimal peak of active memory. Significant gains are obtained compared to Liu's approach. We also study the problem of minimizing the total memory and compare various new schemes. A number of experiments shows the interest of these approaches. Résumé Nous nous intéressonsà l'optimisation mémoire pour l'approche multi-frontale. Repartant des algorithmes proposés par Liu, nous proposons de nouveaux algorithmes et parcours d'arbre visantà minimiser la mémoire. Dans le cas out-of-core nous proposons un algorithme optimal pour la taille de la mémoire active, alors que pour le cadre in-core nous nous in-téressonsà la minimisation de la mémoire totale et proposons plusieurs nouvelles approches. Cetté etude théorique est complétée par des expé-rimentations sur un grand nombre deprobì emes tests qui montrent les améliorations obtenues. Optimal memory minimization algorithms for the multifrontal method
منابع مشابه
Optimal Control Pressure for Leakage Minimization in Water Distribution Systems Using Imperialist Competitive Algorithm
One of the key factors affecting leakage in water distribution systems is network pressure management by putting Pressure Reduce Valves (PRV) in the flow path and optimal regulation of these vales in water networks. This study aimed at investigating optimal pressure management problems so as to minimize leakage in water distribution networks. To do so, an approach was proposed for both optimal ...
متن کاملA CPU-GPU hybrid approach for the unsymmetric multifrontal method
Multifrontal is an efficient direct method for solving large-scale sparse and unsymmetric linear systems. The method transforms a large sparse matrix factorization process into a sequence of factorizations involving smaller dense frontal matrices. Some of these dense operations can be accelerated by using a graphic processing unit (GPU). We analyze the unsymmetricmultifrontalmethod fromboth an ...
متن کاملParallel Computation of Finite Element Navier-Stokes codes using MUMPS Solver
The study deals with the parallelization of 2D and 3D finite element based Navier-Stokes codes using direct solvers. Development of sparse direct solvers using multifrontal solvers has significantly reduced the computational time of direct solution methods. Although limited by its stringent memory requirements, multifrontal solvers can be computationally efficient. First the performance of MUlt...
متن کاملOn the I/O Volume in Out-of-Core Multifrontal Methods with a Flexible Allocation Scheme
Sparse direct solvers, and in particular multifrontal methods, are widely used in various simulation problems. Because of their large memory requirements, the use of out-of-core solvers is compulsory for large-scale problems, where disks are used to extend the core memory. This study is at the junction of two previous independent works: it extends the problem of the minimization of the volume o...
متن کاملRobust Memory-Aware Mappings for Parallel Multifrontal Factorizations
We focus on memory scalability issues in multifrontal solvers like MUMPS. We illustrate why commonly used mapping strategies (e.g., a proportional mapping) cannot achieve a high memory efficiency. We propose a class of “memory-aware” algorithms that aim at maximizing performance under memory constraints. These algorithms provide both accurate memory predictions and a robust solver. We illustrat...
متن کاملذخیره 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 2006