scientific article; zbMATH DE number 3876584
[For the entire collection see Zbl 0527.00033.] The hypercube and its derivative networks (e.g., the shuffle-exchange graph and the cube-connected cycles) have long been known to be powerful networks for parallel computation. Recently, an entirely different class of networks has been discovered that appears to rival the hypercube-based networks in usefulness. In this paper, we describe the new networks (which we call meshes of trees), and we show how they can be used to solve problems such as sorting, matrix-vector multiplication, discrete Fourier transform, transitive closure, minimum spanning tree, integer multiplication and matrix multiplication in \(O(\log N)\) or \(O(\log^ 2 N)\) steps.
- Efficient parallel computations on the reduced mesh of trees organization
- Solving tree problems on a mesh-connected processor array
- scientific article; zbMATH DE number 742956
- Parallel implementation of tree skeletons
- scientific article; zbMATH DE number 2100400
- Parallel algorithms for tree accumulations
- Efficient parallelization of spatial approximation trees
- Tree-based parallel algorithm design
- scientific article; zbMATH DE number 3843145
- Time-optimal tree computations on sparse meshes
- A compact layout for the three-dimensional tree of meshes
- A class of problems efficiently solvable on mesh-connected computers including dynamic expression evaluation
- Product-shuffle networks: Toward reconciling shuffles and butterflies
- Efficient parallel computations on the reduced mesh of trees organization
- Parallel matrix product algorithm in the de Bruijn network using emulation of meshes of trees
- Adapting shuffle-exchange like parallel processing organizations to work as systolic arrays
- Parallel prefix computation on extended multi-mesh network.
- Parallel tree contraction and prefix computations on a large family of interconnection topologies
- The balanced binary tree technique on mesh-connected computers
- Successive approximation in parallel graph algorithms (extended abstract)
- A theory of decomposition into prime factors of layered interconnection networks
- On the performance of parallel matrix factorisation on the hypermesh
- Successive approximation in parallel graph algorithms
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3341891)