Simultaneous Computational and Data Load Balancing in Distributed-Memory Setting

DOI10.1137/22M1485772MaRDI QIDQ5058287zbMATH OpenOpenAlexFDO

Authors Mestan Firat Çeliktuğ, M. Ozan Karsavuran, Seher Acer, Cevdet Aykanat

Publication date 20 December 2022

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/22m1485772

zbMATH Keywords

mesh partitioninghypergraph partitioningdistributed-memory systemsrecursive bipartitioningdata load balancecomputational load balancegeneral sparse matrix-matrix multiplicationmulti-constraint partitioning

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Graph algorithms (graph-theoretic aspects) (05C85) Graph theory (including graph drawing) in computer science (68R10) Hypergraphs (05C65) Edge subsets with special properties (factorization, matching, partitioning, covering and packing, etc.) (05C70)

Recommendations

• New challenges in dynamic load balancing
• A computation + communication load balanced loop partitioning method for distributed memory systems
• Partitioning and dynamic load balancing for the numerical solution of partial differential equations
• A general matrix iterative model for dynamic load balancing
• Parallel static and dynamic multi‐constraint graph partitioning

Cites work

• A Bi-Criteria FPTAS for Scheduling with Memory Constraints on Graphs with Bounded Tree-Width
• A Two-Dimensional Data Distribution Method for Parallel Sparse Matrix-Vector Multiplication
• Adaptive decomposition and remapping algorithms for object-space-parallel direct volume rendering of unstructured grids
• Benchmarking optimization software with performance profiles.
• Communication balancing in parallel sparse matrix-vector multiplication
• Encapsulating Multiple Communication-Cost Metrics in Partitioning Sparse Rectangular Matrices for Parallel Matrix-Vector Multiplies
• Exposing fine-grained parallelism in algebraic multigrid methods
• FPT approximation algorithm for scheduling with memory constraints
• New challenges in dynamic load balancing
• On two-dimensional sparse matrix partitioning: models, methods, and a recipe
• Parallel Scientific Computation
• Simultaneous input and output matrix partitioning for outer-product -- parallel sparse matrix-matrix multiplication
• The University of Florida sparse matrix collection
• UMPa: A multi-objective, multi-level partitioner for communication minimization

Cited in

(2)

• DistDLB: improving cosmology SAMR simulations on distributed computing systems through hierarchical load balancing
• Parallel Processing and Applied Mathematics

Describes a project that uses

Uses Software

• UMPa
• PaToH
• SparseMatrix
• DRAMA

This page was built for publication: Simultaneous Computational and Data Load Balancing in Distributed-Memory Setting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5058287)