Graph Partitioning and Parallel Solvers: Has the Emperor No Clothes? ?
نویسنده
چکیده
Sparse matrix-vector multiplication is the kernel for many scientiic computations. Parallelizing this operation requires the matrix to be divided among processors. This division is commonly phrased in terms of graph partitioning. Although this abstraction has proved to be very useful, it has signiicant aws and limitations. The cost model implicit in this abstraction is only a weak approximation to the true cost of the parallel matrix-vector multiplication. And the graph model is unnecessarily restrictive. This paper will detail the shortcomings of the current paradigm and suggest directions for improvement and further research.
منابع مشابه
Partitioning and Load Balancing for Emerging Parallel Applications and Architectures
An important component of parallel scientific computing is partitioning – the assignment of work to processors. This assignment occurs at the start of a computation (“static” partitioning). Often, reassignment also is done during a computation (“dynamic” partitioning) to redistribute work as the computation changes. The goal of partitioning is to assign work to processors in a way that minimize...
متن کاملCombinatorial Problems in High-Performance Computing: Partitioning
Partitioning is of fundamental importance in high-performance computing: partitioning the data and the associated computational work in an optimal manner leads to good load balance and minimal communication in parallel computations on modern architectures. Often, the computation is irregular and the data set is described by a sparse matrix, a graph, or a hypergraph. This results in a combinator...
متن کاملParallel partitioning with Zoltan: Is hypergraph partitioning worth it?
Graph partitioning is an important and well studied problem in combinatorial scientific computing, and is commonly used to reduce communication in parallel computing. Different models (graph, hypergraph) and objectives (edge cut, boundary vertices) have been proposed. Hypergraph partitioning has become increasingly popular over the last decade. Its main strength is that it accurately captures c...
متن کاملEngineering Graph Partitioning Algorithms
The paper gives an overview of our recent work on balanced graph partitioning – partition the nodes of a graph into k blocks such that all blocks have approximately equal size and such that the number of cut edges is small. This problem has numerous applications for example in parallel processing. We report on a scalable parallelization and a number of improvements on the classical multilevel a...
متن کاملRelaxation-Based Coarsening for Multilevel Hypergraph Partitioning
Multilevel partitioning methods that are inspired by principles of multiscaling are the most powerful practical hypergraph partitioning solvers. Hypergraph partitioning has many applications in disciplines ranging from scientific computing to data science. In this paper we introduce the concept of algebraic distance on hypergraphs and demonstrate its use as an algorithmic component in the coars...
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 1998