Poster (After Call)

FZJ-2022-02583

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png beNNch – Finding Performance Bottlenecks of Neuronal Network Simulators

Albers, J. (Corresponding author)FZJ* ; Pronold, J.FZJ* ; Kurth, A.FZJ* ; Vennemo, S. B. ; Haghighi Mood, K.FZJ* ; Patronis, A. ; Terhorst, D.FZJ* ; Jordan, J. ; Kunkel, S. ; Tetzlaff, T.FZJ* ; Diesmann, M.FZJ* ; Senk, J.FZJ*

2022

Please use a persistent id in citations: http://hdl.handle.net/2128/31614

Abstract: Modern computational neuroscience seeks to explain the dynamics and function of the brain by constructing models with ever more biological detail. This can, for example, take the form of sophisticated connectivity schemes [1] or involve the simultaneous simulation of multiple brain areas [2]. To enable progress in these studies, the simulation of models needs to become faster, calling for more efficient implementations of the underlying simulators. Performance benchmark- ing guides software development since it is hard to predict the impact of algorithm adaptations on the performance of complex software such as neuronal network simulators [3]. The particular challenge for these simulators is that executing benchmarks naturally involves the simulation of a diverse range of network models as they may uncover different performance limitations due to their variation in size, synaptic density and distribution of delays [4]. In addition, maintain- ing an accessible library of past results while keeping track of metadata that specifies hardware, software, simulator and model configurations is a difficult task. Here, we introduce beNNch [5] – a recently developed framework for benchmarking neuronal network simulations – and walk through a typical use case, highlighting how it simplifies workflows and enables sustainable use of computing resources.[1] Billeh, Y. N., Cai, B., Gratiy, S. L., Dai, K., Iyer, R., Gouwens, N. W., et al. (2020). System- atic Integration of Structural and Functional Data into Multi-scale Models of Mouse Primary Visual Cortex. Neuron 106, 388-403.e18. doi: 10.1016/j.neuron.2020.01.040 [2] Schmidt, M., Bakker, R., Hilgetag, C. C., Diesmann, M., and van Albada, S. J. (2018a). Multi-scale ac- count of the network structure of macaque visual cortex. Brain Struct Funct. 223, 1409–1435. doi: 10.1007/s00429-017-1554-4 [3] Jordan, J., Ippen, T., Helias, M., Kitayama, I., Sato, M., Igarashi, J., et al. (2018). Extremely scalable spiking neuronal network simulation code: from laptops to exascale computers. Front. Neuroinform. 12:2. doi: 10.3389/fninf.2018.00002 [4] Albers, J., Pronold, J., Kurth, A. C., Vennemo, S. B., Haghighi Mood, K., Patronis, A., et al. (in press). A Modular Workflow for Performance Benchmarking of Neuronal Network Simulations. Front. Neuroinform. doi: 10.3389/fninf.2022.837549 [5] https://github.com/INM-6/beNNch

---

Contributing Institute(s):

1. Computational and Systems Neuroscience (INM-6)
2. Theoretical Neuroscience (IAS-6)
3. Jara-Institut Brain structure-function relationships (INM-10)

Research Program(s):

1. 5234 - Emerging NC Architectures (POF4-523) (POF4-523)
2. HBP SGA3 - Human Brain Project Specific Grant Agreement 3 (945539) (945539)
3. DEEP-EST - DEEP - Extreme Scale Technologies (754304) (754304)
4. ACA - Advanced Computing Architectures (SO-092) (SO-092)
5. JL SMHB - Joint Lab Supercomputing and Modeling for the Human Brain (JL SMHB-2021-2027) (JL SMHB-2021-2027)
6. GRK 2416:  MultiSenses-MultiScales: Novel approaches to decipher neural processing in multisensory integration (368482240) (368482240)
7. MetaMoSim - Generic metadata management for reproducible high-performance-computing simulation workflows - MetaMoSim (ZT-I-PF-3-026) (ZT-I-PF-3-026)
8. PhD no Grant - Doktorand ohne besondere Förderung (PHD-NO-GRANT-20170405) (PHD-NO-GRANT-20170405)

Appears in the scientific report 2022

---

Database coverage:

---