Browsing by Keywords "Sparse matrices"
Now showing items 120 of 23

Approximate MLFMA as an efficient preconditioner
(IEEE, 2007)In this work, we propose a preconditioner that approximates the dense system operator. For this purpose, we develop an approximate multilevel fast multipole algorithm (AMLFMA), which performs a much faster matrixvector ... 
Cache locality exploiting methods and models for sparse matrixvector multiplication
(Bilkent University, 2009)The sparse matrixvector multiplication (SpMxV) is an important kernel operation widely used in linear solvers. The same sparse matrix is multiplied by a dense vector repeatedly in these solvers to solve a system of ... 
An effective preconditioner based on schur complement reduction for integralequation formulations of dielectric problems
(IEEE, 2009)The author consider effective preconditioning of recently proposed two integralequation formulations for dielectrics; the combined tangential formulation (CTF) and the electric and magnetic current combinedfield integral ... 
Exploiting locality in sparse matrixmatrix multiplication on manycore rchitectures
(IEEE Computer Society, 2017)Exploiting spatial and temporal localities is investigated for efficient rowbyrow parallelization of general sparse matrixmatrix multiplication (SpGEMM) operation of the form C=A,B on manycore architectures. Hypergraph ... 
Hypergraph models for parallel sparse matrixmatrix multiplication
(Bilkent University, 201509)Multiplication of two sparse matrices (i.e., sparse matrixmatrix multiplication, which is abbreviated as SpGEMM) is a widely used kernel in many applications such as molecular dynamics simulations, graph operations, and ... 
A hypergraph partitioning model for profile minimization
(Society for Industrial and Applied Mathematics Publications, 2019)In this paper, the aim is to symmetrically permute the rows and columns of a given sparse symmetric matrix so that the profile of the permuted matrix is minimized. We formulate this permutation problem by first defining ... 
Hypergraphpartitioningbased decomposition for parallel sparsematrix vector multiplication
(IEEE, 1999)In this work, we show that the standard graphpartitioningbased decomposition of sparse matrices does not reflect the actual communication volume requirement for parallel matrixvector multiplication. We propose two ... 
Improving performance of sparse matrix dense matrix multiplication on largescale parallel systems
(Elsevier BV, 2016)We propose a comprehensive and generic framework to minimize multiple and different volumebased communication cost metrics for sparse matrix dense matrix multiplication (SpMM). SpMM is an important kernel that finds ... 
Latencycentric models and methods for scaling sparse operations
(Bilkent University, 201608)Parallelization of sparse kernels and operations on largescale distributed memory systems remains as a major challenge due to everincreasing scale of modern high performance computing systems and multiple con icting ... 
Localityaware parallel sparse matrixvector and matrixtransposevector multiplication on manycore processors
(Institute of Electrical and Electronics Engineers, 2016)Sparse matrixvector and matrixtransposevector multiplication (SpMMTV) repeatedly performed as z ← ATx and y ← A z (or y ← A w) for the same sparse matrix A is a kernel operation widely used in various iterative solvers. ... 
Minimizing communication latencies in conjugate gradient type parallel iterative solvers
(Bilkent University, 200107) 
ON twodimensional sparse matrix partitioning: models, methods, and a recipe
(Society for Industrial and Applied Mathematics, 2010)We consider twodimensional partitioning of general sparse matrices for parallel sparse matrixvector multiply operation. We present three hypergraphpartitioningbased methods, each having unique advantages. The first one ... 
Parallel sparse matrixvector multiplies and iterative solvers
(Bilkent University, 2005)Sparse matrixvector multiply (SpMxV) operations are in the kernel of many scientific computing applications. Therefore, efficient parallelization of SpMxV operations is of prime importance to scientific computing ... 
Parallelization of Sparse Matrix Kernels for big data applications
(Springer, 2016)Analysis of big data on largescale distributed systems often necessitates efficient parallel graph algorithms that are used to explore the relationships between individual components. Graph algorithms use the basic adjacency ... 
Partitioning hypergraphs in scientific computing applications through vertex separators on graphs
(Society for Industrial and Applied Mathematics, 2012)The modeling flexibility provided by hypergraphs has drawn a lot of interest from the combinatorial scientific community, leading to novel models and algorithms, their applications, and development of associated tools. ... 
Partitioning models for scaling distributed graph computations
(Bilkent University, 201909)The focus of this thesis is intelligent partitioning models and methods for scaling the performance of parallel graph computations on distributedmemory systems. Distributed databases utilize graph partitioning to provide ... 
Preconditioning large integralequation problems involving complex targets
(IEEE, 200807)When the target problem is small in terms of the wavelength, simple preconditioners, such as BDP, may sufficiently accelerate the convergence. On the other hand, for largescale problems, the matrix equations become much ... 
Recursive bipartitioning models for performance improvement in sparse matrix computations
(Bilkent University, 201709)Sparse matrix computations are among the most important building blocks of linear algebra and arise in many scienti c and engineering problems. Depending on the problem type, these computations may be in the form of ... 
Reducing communication volume overhead in largescale parallel SpGEMM
(Bilkent University, 201612)Sparse matrixmatrix multiplication of the form of C = A x B, C = A x A and C = A x AT is a key operation in various domains and is characterized with high complexity and runtime overhead. There exist models for parallelizing ... 
Reducing latency cost in 2D sparse matrix partitioning models
(Elsevier BV, 2016)Sparse matrix partitioning is a common technique used for improving performance of parallel linear iterative solvers. Compared to solvers used for symmetric linear systems, solvers for nonsymmetric systems offer more ...