BrainFrame: a node-level heterogeneous accelerator platform for neuron simulations

Smaragdos, Georgios; Chatzikonstantis, Georgios; Kukreja, Rahul; Sidiropoulos, Harry; Rodopoulos, Dimitrios; Sourdis, Ioannis; Al-Ars, Zaid; Kachris, Christoforos; Soudris, Dimitrios; Zeeuw, Chris I. De; Strydis, Christos

doi:10.1088/1741-2552/aa7fc5

Cited by 26 publications

(27 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is shown that when density grows, even more floating-point operations need to be performed to simulate the model. It also reveals that the model is compute bound for larger densities [29]. This makes it an excellent target for HPC fabrics such as the GPU.…”

Section: Multi-node Accelerator Technologymentioning

confidence: 86%

“…The goal of our work is BrainFrame, which is a heterogeneous acceleration platform that incorporates three distinct acceleration technologies, Intel Xeon-Phi CPUs, NVIDIA GP-GPUs, and Maxeler Dataflow Engines (DFEs). So far, we have simulated the ION model on a single-node GPU [7], Xeon-Phi [29], and DFE [30] setup as well as on a multi-node (eight-way) Xeon-Phi [1] setup. Eventually, BrainFrame will move toward multi-node heterogeneity and into the Cloud for all to access.…”

Section: Related Work: Survey Of Neural-network Simulatorsmentioning

confidence: 99%

“…The larger and more complex a simulation becomes, the more computationally intensive it will be. Complex SNN simulations do not scale linearly with large-scale problem sizes (e.g., >100,000 neurons) [1,7,29], which makes them impractical for researchers. High-Performance Computing (HPC) technology such as Graphics-Processing Units (GPUs) has often been used to cope with this limitation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Exploring Complex Brain-Simulation Workloads on Multi-GPU Deployments

Vlag

Smaragdos

Al-Ars

et al. 2019

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

In-silico brain simulations are the de-facto tools computational neuroscientists use to understand large-scale and complex brain-function dynamics. Current brain simulators do not scale efficiently enough to largescale problem sizes (e.g., >100,000 neurons) when simulating biophysically complex neuron models. The goal of this work is to explore the use of true multi-GPU acceleration through NVIDIA's GPUDirect technology on computationally challenging brain models and to assess their scalability. The brain model used is a state-of-the-art, extended Hodgkin-Huxley, biophysically meaningful, three-compartmental model of the inferior-olivary nucleus. The Hodgkin-Huxley model is the most widely adopted conductance-based neuron representation, and thus the results from simulating this representative workload are relevant for many other brain experiments. Not only the actual network-simulation times but also the network-setup times were taken into account when designing and benchmarking the multi-GPU version, an aspect often ignored in similar previous work. Network sizes varying from 65K to 2M cells, with 10 and 1,000 synapses per neuron were executed on 8, 16, 24, and 32 GPUs. Without loss of generality, simulations were run for 100 ms of biological time. Findings indicate that communication overheads do not dominate overall execution while scaling the network size up is computationally tractable. This scalable design proves that large-network simulations of complex neural models are possible using a multi-GPU design with GPUDirect.

show abstract

Section: Multi-node Accelerator Technologymentioning

confidence: 86%

Section: Related Work: Survey Of Neural-network Simulatorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exploring Complex Brain-Simulation Workloads on Multi-GPU Deployments

Vlag

Smaragdos

Al-Ars

et al. 2019

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

show abstract

“…This work is the first, to our knowledge, to extensively study the scaling differences between two different AdEx network configurations; an externally stimulated network as well as a self-stimulated one. Furthermore, the simulator is part of an effort to develop an acceleration platform for computationally challenging neuroscientific simulations [7].…”

Section: Related Workmentioning

confidence: 99%

GPU Implementation of Neural-Network Simulations Based on Adaptive-Exponential Models

Neofytou

Chatzikonstantis

Magkanaris

et al. 2019

2019 IEEE 19th International Conference on Bioinformatics and Bioengineering (BIBE)

Self Cite

View full text Add to dashboard Cite

Detailed brain modeling has been presenting significant challenges to the world of high-performance computing (HPC), posing computational problems that can benefit from modern hardware-acceleration technologies. We explore the capacity of GPUs for simulating large-scale neuronal networks based on the Adaptive Exponential neuron-model, which is widely used in the neuroscientific community. Our GPU-powered simulator acts as a benchmark to evaluate the strengths and limitations of modern GPUs, as well as to explore their scaling properties when simulating large neural networks. This work presents an optimized GPU implementation that outperforms a reference multicore implementation by 50×, whereas utilizing a dual-GPU configuration can deliver a speedup of 90× for networks of 20, 000 fully interconnected AdEx neurons.

show abstract

“…The simulation of massive dense networks of the neuron model is very challenging, due to the high demand in floating-point operations per simulation cycle. Single-node deployments in multiple platforms [29], include x86-based architectures (Intel Xeon and Intel Xeon Phi), GPUs and MAX DFEs. The simulation will be used during the implementation and the early evaluation of the heterogeneity and multi-node extensions of the EXA2PRO environment.…”

Section: Exa2pro Benchmarksmentioning

confidence: 99%

EXA2PRO programming environment

Soudris

Papadopoulos

Keßler

et al. 2018

Proceedings of the 18th International Conference on Embedded Computer Systems: Architectures, Modeling, and Simulation

Self Cite

View full text Add to dashboard Cite

The EXA2PRO programming environment will integrate a set of tools and methodologies that will allow to systematically address many exascale computing challenges, including performance, performance portability, programmability, abstraction and reusability, fault tolerance and technical debt. The EXA2PRO tool-chain will enable the efficient deployment of applications in exascale computing systems, by integrating high-level software abstractions that offer performance portability and efficient exploitation of exascale systems' heterogeneity, tools for efficient memory management, optimizations based on trade-offs between various metrics and fault-tolerance support. Hence, by addressing various aspects of productivity challenges, EXA2PRO is expected to have significant impact in the transition to exascale computing, as well as impact from the perspective of applications. The evaluation will be based on 4 applications from 4 different domains that will be deployed in JUELICH supercomputing center. The EXA2PRO will generate exploitable results in the form of a tool-chain that support diverse exascale heterogeneous supercomputing centers and concrete improvements in various exascale computing challenges.

show abstract

BrainFrame: a node-level heterogeneous accelerator platform for neuron simulations

Cited by 26 publications

References 27 publications

Exploring Complex Brain-Simulation Workloads on Multi-GPU Deployments

Exploring Complex Brain-Simulation Workloads on Multi-GPU Deployments

GPU Implementation of Neural-Network Simulations Based on Adaptive-Exponential Models

EXA2PRO programming environment

Contact Info

Product

Resources

About