Demonstrating BrainScaleS-2 Inter-Chip Pulse-Communication using EXTOLL

02/24/2022
by   Tobias Thommes, et al.
0

The BrainScaleS-2 (BSS-2) Neuromorphic Computing System currently consists of multiple single-chip setups, which are connected to a compute cluster via Gigabit-Ethernet network technology. This is convenient for small experiments, where the neural networks fit into a single chip. When modeling networks of larger size, neurons have to be connected across chip boundaries. We implement these connections for BSS-2 using the EXTOLL networking technology. This provides high bandwidths and low latencies, as well as high message rates. Here, we describe the targeted pulse-routing implementation and required extensions to the BSS-2 software stack. We as well demonstrate feed-forward pulse-routing on BSS-2 using a scaled-down version without temporal merging.

READ FULL TEXT
research
11/30/2021

BrainScaleS Large Scale Spike Communication using Extoll

The BrainScaleS Neuromorphic Computing System is currently connected to ...
research
09/17/2018

The Impact of On-chip Communication on Memory Technologies for Neuromorphic Systems

Emergent nanoscale non-volatile memory technologies with high integratio...
research
03/05/2020

Predicting Memory Compiler Performance Outputs using Feed-Forward Neural Networks

Typical semiconductor chips include thousands of mostly small memories. ...
research
01/25/2017

Neuromorphic computing with nanoscale spintronic oscillators

Neurons in the brain behave as non-linear oscillators, which develop rhy...
research
12/10/2021

Attacks on Wireless Coexistence: Exploiting Cross-Technology Performance Features for Inter-Chip Privilege Escalation

Modern mobile devices feature multiple wireless technologies, such as Bl...
research
07/18/2017

Logic Programming approaches for routing fault-free and maximally-parallel Wavelength Routed Optical Networks on Chip (Application paper)

One promising trend in digital system integration consists of boosting o...
research
10/20/2020

AutoBSS: An Efficient Algorithm for Block Stacking Style Search

Neural network architecture design mostly focuses on the new convolution...

Please sign up or login with your details

Forgot password? Click here to reset