YPOLOGIST
Programmable, General-Purpose Parallel Accelerator for Heteregonous Systems and Applications
Green Novel Architecture
- Achieves 2-3X energy savings in linear algebra applications compared to off-the-shelf solutions
- Delivers 337% better performance in MH/sec/Watt compared to NVIDIA GPUs
- Provides up to 30x speed improvement and 300x energy efficiency in ASIC implementations
11x
Energy Efficiency
3x
Less Area
6x
Acceleration
3x
Less Cost (at least)
*The comparisons are made with the Nvidia chip implemented on the same technology, operating at the same frequency.