Skip to main content

C0-microSD Benchmarks

The benchmarks below evaluate the speedup achieved by replacing Monte Carlo simulation of sensor calibration and conversion routines with UxHw-based computation running on the C0-microSD (Athens-16 representation), versus Monte Carlo executions on an ARM Cortex-M33 microcontroller (150 MHz, ~39mW).

The application kernels in the table correspond to the manufacturer-specified calibration algorithms that convert each measured ADC value into a calibrated sensor output.

The number of Monte Carlo iterations for each application is chosen such that the output probability distribution reaches the same degree of convergence as the UxHw Athens-16 result.

The following figures come from the Signaloid-Compute-Module-Demo-Sensor demo.

ApplicationMonte Carlo iterationsMonte Carlo latencyUxHw latencySpeedup
Flusso FLS110291k3 s1.63 s1.8×
NXP MPX4100A185k0.7 s0.42 s1.6×
NXP MPXx6250A215k0.8 s0.42 s1.9×
Sensirion SDP3x193k2.3 s0.42 s5.5×
Sensirion SFM31002.7M8.2 s23.56 ms350×
Sensirion SHT3xARP193k1.6 s1.26 s1.3×
Sensirion SHT4xI204k1.7 s1.26 s1.4×

UxHw latency is the latency for reaching the same degree of convergence of the distributional result as the multi-iteration Monte Carlo.

Next steps