Our Technology
Aspinity’s AnalogML™ IP delivers the most efficient accelerator architecture for AI computing.
Digital Accelerator | Analog In-Memory-Computing (IMC) | Aspinity AnalogML™ | |
---|---|---|---|
Data Movement | HighContinual movement of instructions, parameters, and intermediate data to/from cache | LowParameters are embedded; still some movement of instructions and intermediate data to/from cache | LowParameters and configuration (i.e., instructions) are embedded; Signals stream through |
Parameter Precision | HighHigh-precision datatypes often supported | Low-MediumLimited to ≤ 8 bits precision, much less in many implementations | High10+ bit precision |
MAC Precision | HighHigh-precision datatypes often supported | Low“Memory as compute” precision limited by linearity, noise, and mismatch | HighParameters co-located with compute allows linear multiplication circuits to be used |
MAC Efficiency | Low1000’s of transistors and transfer instructions and operands from memory | Medium Efficient multiplication in analog; still fetch input vector from memory | HighEfficient multiplication in analog |
NN Efficiency | LowData movement to/from memory | MediumMultiply is in memory but activation requires digital | HighMAC & activation all within analog circuitry |
Robustness to Temperature & Manufacturing Variation | HighLimited issues with digital circuitry | LowChallenged with long signal chains | HighSilicon-proven and dynamic approach to on-the-fly trimming |
Precision Analog Parameters
High Performance Acceleration
Aspinity patented analog memory is co-located in circuits
10+ bits precision for storing parameters (weights, etc.)
Permanent and accurate parameter storage (no memory fetch)
Weight quantization not required to save power
Most efficient inferencing with AnalogML
- Fully analog data path is inherently more efficient than digital.
- Analog multiply accumulate functions use <10 transistors (compared to 1000’s in digital).
- All-analog NN layer includes MAC and activation functions (no digital conversions required).
- Memory-at-compute eliminates time and energy spent fetching data.
- Intermediate values not stored in memory – all signals are streamed through layer-to-layer as analog signals.
- Scalable neural network architecture addresses wide range of edge inferencing applications
AnalogML enables robust and accurate immunity to analog variability
- Dynamic software-driven solution for on-the-fly variation trimming
- Leverages high-precision analog parameters for localized fine tuning/trimming
- Not susceptible to environmental or manufacturing variations
- Repeatable and consistent analog computing platform can be scaled to larger neural networks and smaller process nodes to meet the needs of next generation AI computing
Modern, easy-to-use Python based training and execution
Integration with common
ML frameworks
Simple and quick compilation
to AnalogML hardware
Supports CNN, RNN, DNN, and other model types