Developing Data Redistribution Algorithms
Purpose
[edit | edit source]- Developing Communication Models that can represent the characteristics about each communication link and its bandwidth.
- Developing Data Redistribution Algorithms using AI techniques.
- Verifying Algorithms by using NOC SystemC Simulators.
Idea Sketch
[edit | edit source]- Idea.1.A (pdf)
Communication Patterns and Models
[edit | edit source]Parallel Prefix Sum
[edit | edit source]
FFT
[edit | edit source]
Redistribution Algorithms using a Generalized Circulant Matrix
[edit | edit source]Applying Sudoku Problems to Data Redistribution
[edit | edit source]Sudoku Codes in Haskell
[edit | edit source]from [ Haskell_programming_in_plain_view ]
- Sudoku Background (pdf)
- Bird's Implementation
Sudoku Codes in Prolog
[edit | edit source]
Finding relationship with other algorithms
[edit | edit source]FFT Algorithms using a Tensor Product
[edit | edit source]
FFT Algorithms using a Mixed Radix
[edit | edit source]
Background
[edit | edit source]- Y. Ishikawa : "Efficient Collective Operations for Clusters in Long-and-Fast Networks"
- R. V. de Geijn : "Collective Communications on Architectures that Support Simultaneous Communications over Multiple Links"
- R. Rabenseifner : "Optimization of Collective Communication Operations in MPICH"
- NoC Simulator
- - NIRGAM (SystemC)- University of Southampton, 2007
- - OCCN (SystemC) - ST Microelectronics, 2005
- - TOPAZ (C++) - University of Cantabria, 2012
- - HNoCs (OMNeT++) - Technion, 2011
- MPI+NOC Literature to be surveyed.
- Y. W. Lim : "Efficient algorithms for block-cyclic redistribution of arrays", 1999
- Circulant Matrix based redistribution (pdf)
go to [ Electrical_&_Computer_Engineering_Studies ]
* Idea.2.A (pdf) moved to CORDIC Hardware Implementations