A novel simulator for extended Hodgkin-Huxley neural networks

Panagiotou, Sotirios; Miedema, Rene; Sidiropoulos, Harry; Smaragdos, Georgios; Strydis, Christos; Soudris, Dimitrios

doi:10.1109/bibe50027.2020.00071

Cited by 5 publications

(2 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One last categorization can be made according to the target hardware platform of the simulator system. A wide variety of hardware has been employed, starting from regular CPUs [12], [15], [32], [36], [40]- [42], GPUs [1], [5]- [7], [13], [14], [16], [20], [27], [30], [33]- [35], [43], [44], combination of CPUs and GPUs [19], [38], [39], [45], and including even custombuilt hardware [11], [20], [21], [25], [26], [46]. Recent simulators are increasingly adopting the GPU architecture.…”

Section: A Snn Simulator Classificationmentioning

confidence: 99%

Even Faster SNN Simulation with Lazy+Event-driven Plasticity and Shared Atomics

Bautembach,

Oikonomidis,

Argyros

2021

Preprint

View full text Add to dashboard Cite

We present two novel optimizations that accelerate clock-based spiking neural network (SNN) simulators. The first one targets spike timing dependent plasticity (STDP). It combines lazy-with event-driven plasticity and efficiently facilitates the computation of pre-and post-synaptic spikes using bitfields and integer intrinsics. It offers higher bandwidth than eventdriven plasticity alone and achieves a 1.5×-2× speedup over our closest competitor. The second optimization targets spike delivery. We partition our graph representation in a way that bounds the number of neurons that need be updated at any given time which allows us to perform said update in shared memory instead of global memory. This is 2×-2.5× faster than our closest competitor. Both optimizations represent the final evolutionary stages of years of iteration on STDP and spike delivery inside "Spice" (/spaIk/), our state of the art SNN simulator. The proposed optimizations are not exclusive to our graph representation or pipeline but are applicable to a multitude of simulator designs. We evaluate our performance on three wellestablished models and compare ourselves against three other state of the art simulators.

show abstract

Section: A Snn Simulator Classificationmentioning

confidence: 99%

Even Faster SNN Simulation with Lazy+Event-driven Plasticity and Shared Atomics

Bautembach,

Oikonomidis,

Argyros

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Lastly, simulators can also be classified according to the employed hardware platform. Common targets include regular CPUs [10], [27], [32], [34], [35], [38], [39], GPUs [2], [3], [8], 1. We store our network's topology in a padded adjacency list (a).…”

Section: A Simulator Classificationsmentioning

confidence: 99%

Multi-GPU SNN Simulation with Static Load Balancing

Bautembach,

Oikonomidis,

Argyros

2021

Preprint

View full text Add to dashboard Cite

We present a SNN simulator which scales to millions of neurons, billions of synapses, and 8 GPUs. This is made possible by 1) a novel, cache-aware spike transmission algorithm 2) a model parallel multi-GPU distribution scheme and 3) a static, yet very effective load balancing strategy. The simulator further features an easy to use API and the ability to create custom models. We compare the proposed simulator against two state of the art ones on a series of benchmarks using three wellestablished models. We find that our simulator is faster, consumes less memory, and scales linearly with the number of GPUs.

show abstract