HOME
UP
  PREV
|ENDOFPACK|
HLS to replace Von Neumann?
Certain maniacs predict FPGA may replace Von Neumann!
Spatio-Parallel processing uses less energy than equivalent temporal processing (ie at higher clock rates) for various reasons. David Greaves gives nine:
- Pollack's rule states that energy use in a Von Neumann CPU grows with square of its IPC. But the FPGA with a static schedule moves the out-of-order overheads to compile time.
- To clock CMOS at a higher frequency needs a higher voltage, so energy use has quadratic growth with frequency.
- Von Neumann SIMD extensions greatly amortise fetch and decode energy, but FPGA does better, supporting precise custom word widths, so no waste at all.
- FPGA can implement massively-fused accumulate rather than re-normalising after each summation.
- Memory bandwidth: FPGA has always had superb on-chip memory bandwidth but latest generation FPGA exceeds CPU on DRAM bandwidth too.
- FPGA using combinational logic uses zero energy re-computing sub-expressions whose support has not changed. And it has no overhead determining whether it has changed.
- FPGA has zero conventional instruction fetch and decode energy and its controlling micro-sequencer or predication energy can be close to zero.
- Data locality can easily be exploited on FPGA --- operands are held closer to ALUs, giving near-data-processing (but the FPGA overall size is x10 times larger (x100 area) owing to overhead of making it reconfigurable).
- The massively-parallel premise of the FPGA is the correct way forward, as indicated by asymptotic limit studies [DeHon].