View Item 
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        •   Utrecht University Student Theses Repository Home
        • UU Theses Repository
        • Theses
        • View Item
        JavaScript is disabled for your browser. Some features of this site may not work without it.

        Browse

        All of UU Student Theses RepositoryBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

        Optimal matrix distribution by simulated annealing for a parallel sparse matrix-vector multiplication

        Thumbnail
        View/Open
        Thesis_Final_Version.pdf (662.6Kb)
        Publication date
        2025
        Author
        Noordam, Pieter
        Metadata
        Show full item record
        Summary
        In parallel Sparse Matrix-Vector Multiplication (SpMV), finding an efficient matrix partitioning is crucial to minimize communication cost. This thesis explores the use of Simulated Annealing to find optimal matrix bipartitions. The method uses row/column-based moves to explore the possible partitioning solutions. Our findings indicate that the introduced method is promising for bipartitioning large sparse matrices, especially when exact algorithms prove to be computationally expensive.
        URI
        https://studenttheses.uu.nl/handle/20.500.12932/48775
        Collections
        • Theses
        Utrecht university logo