Scalable Construction of Spiking Neural Networks using up to thousands of GPUs archive
Authors/Creators
-
Tiddia, Gianmarco
(Contact person)1
-
Villamar, José
(Data manager)2
-
Golosio, Bruno
(Project leader)3
-
Pontisso, Luca
(Project member)4
-
Simula, Francesco
(Project member)5
-
Babu, Pooja
(Project member)2
-
Pastorelli, Elena
(Project member)6
-
Morrison, Abigail
(Project member)2
-
Diesmann, Markus
(Project member)2
-
Lonardo, Alessandro
(Project member)4
-
Paolucci, Pier Stanislao
(Project member)6
-
Senk, Johanna
(Project manager)7, 2
-
1.
Istituto Nazionale di Fisica Nucleare, Sezione di Cagliari
-
2.
Forschungszentrum Jülich
-
3.
University of Cagliari
-
4.
Istituto Nazionale di Fisica Nucleare, Sezione di Roma I
- 5. Istituto Nazionale di Fisica Nucleare Sezione di Roma
-
6.
Istituto Nazionale di Fisica Nucleare
-
7.
University of Sussex
Description
Diverse scientific and engineering research areas deal with discrete, time-stamped changes in large systems of interacting delay differential equations. Simulating such complex systems at scale on high-performance computing clusters demands efficient management of communication and memory. Inspired by the human cerebral cortex — a sparsely connected network of O(10^10) neurons, each forming O(10^3)--O(10^4) synapses and communicating via short electrical pulses called spikes — we study the simulation of large-scale spiking neural networks for computational neuroscience research. This work presents a novel network construction method for multi-GPU clusters and upcoming exascale supercomputers using the Message Passing Interface (MPI), where each process builds its local connectivity and prepares the data structures for efficient spike exchange across the cluster during state propagation. We demonstrate scaling performance of two cortical models using point-to-point and collective communication, respectively.
Files
ngpu_scalable_network_construction.zip
Files
(322.8 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:1848549293a65405459de8186515d8cc
|
322.8 MB | Preview Download |
Additional details
Related works
- Continues
- Software: 10.5281/zenodo.7744239 (DOI)
Funding
- European Commission
- HBP SGA3 - Human Brain Project Specific Grant Agreement 3 945539
- European Commission
- EBRAINS 2.0 - EBRAINS 2.0: A Research Infrastructure to Advance Neuroscience and Brain Health 101147319
- Ministero dell'università e della ricerca
- PNRR project PE0000013-FAIR, funded by NextGenerationEU PE0000013-FAIR — CUP I53C22001400006
- Ministero dell'università e della ricerca
- e.INS – Ecosystem of Innovation for Next Generation Sardinia, spoke 10 (PNRR) ECS00000038 — CUP F53C22000430001
Software
- Repository URL
- https://github.com/gmtiddia/ngpu_scalable_network_construction
- Programming language
- Python
- Development Status
- Active