Power and Performance Tradeoff of a Floating-Point Intensive Kernel on OpenCL FPGA Platform.
Zheming JinHal FinkelPublished in: IPDPS Workshops (2018)
Keyphrases
- floating point
- graphics processing units
- real time
- fixed point
- power consumption
- square root
- field programmable gate array
- reconfigurable hardware
- shared memory
- parallel architecture
- floating point unit
- sparse matrices
- low cost
- power reduction
- kernel function
- parallel algorithm
- data structure
- fast fourier transform
- floating point arithmetic
- instruction set
- communication protocol
- hardware implementation
- image processing
- interval arithmetic
- computer systems
- higher order
- dynamic programming