• Corpus ID: 218889474

On Optimal Partitioning For Sparse Matrices In Variable Block Row Format

@article{Ahrens2020OnOP,
  title={On Optimal Partitioning For Sparse Matrices In Variable Block Row Format},
  author={Peter Ahrens and Erik G. Boman},
  journal={ArXiv},
  year={2020},
  volume={abs/2005.12414}
}
The Variable Block Row (VBR) format is an influential blocked sparse matrix format designed to represent shared sparsity structure between adjacent rows and columns. VBR consists of groups of adjacent rows and columns, storing the resulting blocks that contain nonzeros in a dense format. This reduces the memory footprint and enables optimizations such as register blocking and instruction-level parallelism. Existing approaches use heuristics to determine which rows and columns should be grouped… 

Figures and Tables from this paper

Load Plus Communication Balancing of Contiguous Sparse Matrix Partitions.
TLDR
This work considers contiguous partitions, where the rows (or columns) of a sparse matrix with nonzeros are split into $K$ parts without reordering, and proposes exact and approximate contiguous partitioners that minimize the maximum runtime of any processor under a diverse family of cost models.
Load Plus Communication Balancing in Contiguous Partitions for Distributed Sparse Matrices: Linear-Time Algorithms
TLDR
This work presents exact and approximate contiguous partitioning algorithms that minimize the runtime of the longest-running processor under cost models that combine work factors and hypergraph communication factors.
Contiguous Graph Partitioning For Optimal Total Or Bottleneck Communication
TLDR
This work proposes the first near-linear time algorithms for several graph partitioning problems in the contiguous regime, and proposes a new bottleneck cost which reflects the sum of communication and computation on each part.

References

SHOWING 1-10 OF 60 REFERENCES
Perfomance Models for Blocked Sparse Matrix-Vector Multiplication Kernels
TLDR
This paper studies and evaluates a number of representative blocking storage formats and presents a performance model that can accurately select the most suitable blocking storage format and the corresponding block shape and size for a specific sparse matrix.
Julia: A Fresh Approach to Numerical Computing
TLDR
The Julia programming language and its design is introduced---a dance between specialization and abstraction, which recognizes what remains the same after computation, and which is best left untouched as they have been built by the experts.
Automatic performance tuning of sparse matrix kernels
TLDR
An automated system to generate highly efficient, platform-adapted implementations of sparse matrix kernels, and extends SPARSITY to support tuning for a variety of common non-zero patterns arising in practice, and for additional kernels like sparse triangular solve (SpTS) and computation of ATA·x and A ρ·x.
The university of Florida sparse matrix collection
TLDR
The University of Florida Sparse Matrix Collection, a large and actively growing set of sparse matrices that arise in real applications, is described and a new multilevel coarsening scheme is proposed to facilitate this task.
The DynB Sparse Matrix Format Using Variable Sized 2D Blocks for Efficient Sparse Matrix Vector Multiplications with General Matrix Structures
TLDR
Results show that the benefit of blocking formats depend – as to be expected – on the structure of the matrix and that variable sized block formats like DynB can have advantages over fixed size formats and deliver good performance results even for general sparse matrices.
Variable-Sized Blocks for Locality-Aware SpMV
TLDR
This work proposes an analytical model to determine the effective block size for highly irregular sparse matrices by factoring the distribution of non-zeros in the sparse dataset, and generates blocks that are variable-sized as opposed to constant-sized in most existing SpMV algorithms.
Performance Portable Supernode-based Sparse Triangular Solver for Manycore Architectures
TLDR
A sparse triangular solver that takes advantage of the supernodal structures of the triangular matrices that come from the direct factorization of a sparse matrix and is portable to different manycore architectures.
Heuristics for Symmetric Rectilinear Matrix Partitioning
TLDR
This work proposes five heuristics to solve two different variants of the problem of symmetric rectilinear partitioning of square matrices, and presents a thorough experimental evaluation showing the effectiveness of the proposed algorithms.
Heuristics for Symmetric Rectilinear Matrix Partitioning
TLDR
This work proposes five heuristics to solve two different variants of the problem of symmetric rectilinear partitioning of square matrices, and presents a thorough experimental evaluation showing the effectiveness of the proposed algorithms.
...
...