An experimental study of reduced-voltage operation in modern FPGAs for neural network acceleration
We empirically evaluate an undervolting technique, ie, underscaling the circuit supply
voltage below the nominal level, to improve the power-efficiency of Convolutional Neural …
voltage below the nominal level, to improve the power-efficiency of Convolutional Neural …
Revisiting row hammer: A deep dive into understanding and resolving the issue
H Wang, X Peng, Z Liu, X Huang, T Li, B Yang… - Microelectronics …, 2024 - Elsevier
Row hammer is a vulnerability in Dynamic Random Access Memory (DRAM) chips, whereby
repeatedly accessing a specific row in the DRAM chip may cause bit flips in memory cells …
repeatedly accessing a specific row in the DRAM chip may cause bit flips in memory cells …
Burstlink: Techniques for energy-efficient video display for conventional and virtual reality systems
Conventional planar video streaming is the most popular application in mobile systems. The
rapid growth of 360° video content and virtual reality (VR) devices is accelerating the …
rapid growth of 360° video content and virtual reality (VR) devices is accelerating the …
Understanding Timing Error Characteristics from Overclocked Systolic Multiply–Accumulate Arrays in FPGAs
A Chamberlin, A Gerber, M Palmer, T Goodale… - Journal of Low Power …, 2024 - mdpi.com
Artificial Intelligence (AI) hardware accelerators have seen tremendous developments in
recent years due to the rapid growth of AI in multiple fields. Many such accelerators …
recent years due to the rapid growth of AI in multiple fields. Many such accelerators …
Aggressive undervolting of FPGAs: power & reliability trade-offs
B Salami - 2018 - upcommons.upc.edu
In this work, we evaluate aggressive undervolting, ie, voltage underscaling below the
nominal level to reduce the energy consumption of Field Programmable Gate Arrays …
nominal level to reduce the energy consumption of Field Programmable Gate Arrays …
On the Impact of Device-Level Techniques on Energy-Efficiency of Neural Network Accelerators
SM Nabavinejad, B Salami - arXiv preprint arXiv:2106.14079, 2021 - arxiv.org
Energy-efficiency is a key concern for neural network applications. To alleviate this issue,
hardware acceleration using FPGAs or GPUs can provide better energy-efficiency than …
hardware acceleration using FPGAs or GPUs can provide better energy-efficiency than …
Recent advances in DRAM and flash memory architectures
This article features extended summaries and retrospectives of some of the recent research
done by our group, SAFARI, on (1) understanding, characterizing, and modeling various …
done by our group, SAFARI, on (1) understanding, characterizing, and modeling various …
BurstLink: Techniques for Energy-Efficient Conventional and Virtual Reality Video Display
Conventional planar video streaming is the most popular application in mobile systems and
the rapid growth of 360 video content and virtual reality (VR) devices are accelerating the …
the rapid growth of 360 video content and virtual reality (VR) devices are accelerating the …
FPGA'lerde BRAM gerilimi düşürülmesinin güç kazanımı, hata oranı ve sıcaklığın etkileri yönünden araştırılması
F Ağırnas - 2019 - 193.140.108.196
Yüksek performans, düşük güç tüketimi, tasarım esnekliği sağlaması, basit ve yeniden
yapılandırılabilir tasarım özellikleriyle, FPGA'ler gömülü sistemler, savunma ve uzay …
yapılandırılabilir tasarım özellikleriyle, FPGA'ler gömülü sistemler, savunma ve uzay …