% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@INPROCEEDINGS{Lober:1027274,
author = {Lober, Melissa and Diesmann, Markus and Kunkel, Susanne},
title = {{O}ptimizing communication in brain-scale multi-area model
simulations},
school = {RWTH Aachen},
reportid = {FZJ-2024-03723},
year = {2024},
abstract = {With the field of dedicated hardware for neuronal
simulations growing rapidly over the past years,
conventional hardware still serves as an important reference
benchmark while maintaining more flexibility at potentially
lower cost. Kurth et al. [1] have shown that for a realistic
cortical microcircuit model neuronal simulation technology
for conventional hardware keeps pace with novel computing
architectures, such as SpiNNaker [2], regarding real-time
factor as well as energy efficiency. In this project we
develop and advance simulation technology for spiking neural
networks for conventional computer architectures, thereby
challenging and inspiring novel neuromorphic systems. The
communication of spike events constitutes a major bottleneck
in simulations of brain-scale networks with realistic
connectivity, as for example, the multi-area model of
macaque visual cortex [3]. The model consists of 32 cortical
microcircuits, each representing a downscaled area of the
visual cortex at the resolution of single neurons and
synapses. Such models not only have a dense connectivity
within areas but also between areas. Synaptic transmission
delays within an area can be as short as 0.1 ms and
therefore simulations require frequent spike communication
between compute nodes to maintain causality in the network
dynamics [4]. This poses a challenge to the conventional
round-robin scheme used to distribute neurons uniformly
across compute nodes disregarding the network’s specific
topology.We target this challenge and propose a
structure-aware neuron distribution scheme along with a
novel spike-communication framework that exploits this
approach in order to make communication in large-scale
distributed simulations more efficient. In the
structure-aware neuron distribution scheme, neurons are
placed on the hardware in a way that mimics the network’s
topology. Paired with a communication framework that
distinguishes local short delay intra-area communication and
global long delay inter-area communication, the
structure-aware approach minimizes the costly global
communication and thereby reduces simulation time. Our
prototype implementation is fully tested and was developed
within the neuronal simulator tool NEST [5].For the
benchmarking of our approach, we developed a multi-area
model that resembles the macaque multi-area model in terms
of connectivity and work load, while being more easily
scalable as it retains constant activity levels. We show
that the new strategy significantly reduces communication
time in weak-scaling experiments and the effect increases
with an increasing number of compute nodes.[1] Kurth et al.,
Neuromorph. Comput. Eng., 2, 021001, 2022[2] Furber $\&$
Bogdan, Boston-Delft: now publishers, 2020[3] Schmidt et
al., PLoS Comput Biol, 14(10), 1-38, 2018[4] Morrison $\&$
Diesmann, Springer Berlin Heidelberg, pp 267-278, 2008[5]
Gewaltig $\&$ Diesmann, Scholarpedia 2(4):1430 , 2007},
month = {Jun},
date = {2024-06-03},
organization = {International Conference on
Neuromorphic Computing and Engineering,
Aachen (Germany), 3 Jun 2024 - 6 Jun
2024},
subtyp = {After Call},
cin = {IAS-6 / INM-10},
cid = {I:(DE-Juel1)IAS-6-20130828 / I:(DE-Juel1)INM-10-20170113},
pnm = {5231 - Neuroscientific Foundations (POF4-523) / 5232 -
Computational Principles (POF4-523) / JL SMHB - Joint Lab
Supercomputing and Modeling for the Human Brain (JL
SMHB-2021-2027) / BMBF 03ZU1106CB - NeuroSys:
Algorithm-Hardware Co-Design (Projekt C) - B
(BMBF-03ZU1106CB) / EBRAINS 2.0 - EBRAINS 2.0: A Research
Infrastructure to Advance Neuroscience and Brain Health
(101147319)},
pid = {G:(DE-HGF)POF4-5231 / G:(DE-HGF)POF4-5232 / G:(DE-Juel1)JL
SMHB-2021-2027 / G:(DE-Juel1)BMBF-03ZU1106CB /
G:(EU-Grant)101147319},
typ = {PUB:(DE-HGF)24},
doi = {10.34734/FZJ-2024-03723},
url = {https://juser.fz-juelich.de/record/1027274},
}