A4 Vertaisarvioitu artikkeli konferenssijulkaisussa
Exploring NoC jitter effect on simulation of spiking neural networks
Tekijät: Sergei Dytckov, Sushri Sunita Purohit, Masoud Daneshtalab, Juha Plosila, Hannu Tenhunen
Toimittaja: Waleed W. Smari, Vesna Zeljkovic
Konferenssin vakiintunut nimi: High Performance Computing & Simulation, HPCS 2014
Julkaisuvuosi: 2014
Kokoomateoksen nimi: Proceedings of the 2014 International Conference on High Performance Computing & Simulation (HPCS 2014)
Aloitussivu: 693
Lopetussivu: 696
Sivujen määrä: 4
ISBN: 978-1-4799-5312-7
DOI: https://doi.org/10.1109/HPCSim.2014.6903756
Verkko-osoite: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6903756
Rinnakkaistallenteen osoite: https://research.utu.fi/converis/portal/detail/Publication/2227404
The major bottleneck in simulation of large-scale neural networks is the communication problem due to one-to-many neuron connectivity. Network-on-Chip concept has been proposed to address the problem. This work explores the drawback that is introduced by interconnection networks – a delay jitter. The preliminary experiment is held in the spiking neural network simulator introducing variable communicational delay to the simulation. The performance degradation is reported.
Avainsanat:
jitter, network-on-chip, Neural network, self-organizing map (SOM)
Ladattava julkaisu This is an electronic reprint of the original article. |  |
