Automatic design and partitioning of systolic/wavefront arrays for VLSI
From MaRDI portal
(Redirected from Publication:1104739)
We show how systolic/wavefront arrays can be automatically designed and partitioned to solve problems of arbitrary size. Buffer memory and control of a resulting array is regular and simple, and is generated automatically. Also, the throughput of the arrays is matched with the I/0 speed of the host to which it is to be attached. The approach strongly relies upon classical concepts in signal processing, such as signal flow graphs and state transition functional behaviour. Some illustrative examples are included.
Recommendations
- scientific article; zbMATH DE number 5605534
- Simulation-based design of programmable systolic arrays
- Automatic Design of Area-Efficient Configurable ASIC Cores
- scientific article; zbMATH DE number 1941075
- scientific article; zbMATH DE number 125178
- A systolic array with applications to image processing and wire-routing in VLSI circuits
- Polymorphic arrays: A novel VLSI layout for systolic computers
- The design of optimal planar systolic arrays for matrix multiplication
Cites work
Cited in
(7)- Automatic design and partitioning of systolic/wavefront arrays for VLSI
- COMBINING BACKGROUND MEMORY MANAGEMENT AND REGULAR ARRAY CO-PARTITIONING, ILLUSTRATED ON A FULL MOTION ESTIMATION KERNEL
- scientific article; zbMATH DE number 3968510 (Why is no real title available?)
- scientific article; zbMATH DE number 4078817 (Why is no real title available?)
- scientific article; zbMATH DE number 4082984 (Why is no real title available?)
- scientific article; zbMATH DE number 2217830 (Why is no real title available?)
- Optimal piecewise linear schedules for LSGP- and LPGS-decomposed array processors via quadratic programming
This page was built for publication: Automatic design and partitioning of systolic/wavefront arrays for VLSI
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1104739)