Table 3.
Comparison of power usage of modular attractor network simulations running on SpiNNaker with simulations distributed across enough compute nodes of a Cray XC-30 system to match SpiNNaker simulation time.
| Simulation | SpiNNaker | Cray XC-30 | |||
|---|---|---|---|---|---|
| NHC | time [min] | # chips | Peak CPU power usage [W] | # compute nodes | Peak CPU power usage [W] |
| 4 | 17 | 6 | 6 | 2 | 938 |
| 9 | 50 | 12 | 12 | 2 | 938 |
| 16 | 146 | 21 | 21 | 2 | 938 |
| 4 | 9 | 6 | 6 | 4 | 1875 |
| 9 | 23 | 12 | 12 | 14a | 6563 |
| 16 | 62 | 21 | 21 | 9 | 4219 |
Cray XC-30 power usage is based on the 30 kW power usage of an entire Cray XC-30 compute rack (Cray, 2013). SpiNNaker power usage is based on the 1 W peak power usage of the SpiNNaker chip (Furber et al., 2014).
Top: SpiNNaker simulation times include downloading of learned weights and re-uploading required by current software.
Bottom: Time taken to download learned weights, re-generate and re-upload model to SpiNNaker have been removed.
We are unsure why more supercomputer compute nodes are required to match the SpiNNaker simulation times when NHC = 9 than when NHC = 16. We assume this is an artifact of the different scaling properties of the two simulators, but further investigation is outside of the scope of this work.