Results 51 to 60 of about 1,316 (216)

Data Parallel Address Architecture [PDF]

IEEE Computer Architecture Letters, 2006
Data parallel memory systems must maintain a large number of outstanding memory references to fully use increasing DRAM bandwidth in the presence of increasing latency. At the same time, the throughput of modern DRAMs is very sensitive to access pattern's due to the time required to precharge and activate banks and to switch between read and write ...
Jung Ho Ahn, William J. Dally
openaire   +2 more sources

Performance–Complexity Trade‐Offs in Battery Lifetime Prediction with Task‐Aware Transformers

Advanced Science, EarlyView.
FAST‐BatPro integrates convolutional feature extraction, flash Attention, and sparse attention for efficient battery lifetime prediction. Using limited early‐cycle data across multiple chemistries and operating conditions, it achieves robust accuracy while reducing inference latency, computational cost, and energy consumption.
Jingyuan Zhao, Misheng Cai, Zhenghong Wang, Yan Wang, Yuqi Li, Bo Dong, Hewu Wang, Andrew F. Burke, Simona Onori, Stephen J. Harris   +9 more
wiley   +1 more source

SCALABLE PARALLEL TRAINING OF DEEP NEURAL NETWORKS FOR IMAGE CLASSIFICATION USING TENSOR PROCESSING UNITS

Компютерні системи та інформаційні технології
Image classification using machine learning techniques is crucial in fields such as medicine, ecology, and agriculture, where large datasets of images need to be processed efficiently.
Леся МОЧУРАД, Христина ДОЛИНСЬКА, Тетяна УФІМЦЕВА   +2 more
doaj   +1 more source

LC/DC: Lockless Containers and Data Concurrency a Novel Nonblocking Container Library for Multicore Applications

IEEE Access, 2013
Exploiting the parallelism in multiprocessor systems is a major challenge in modern computer science. Multicore programming demands a change in the way we design and use fundamental data structures.
Damian Dechev, Pierre Laborde, Steven D. Feldman   +2 more
doaj   +1 more source

Neuromorphic Near‐Sensor and In‐Sensor Computing Enabled by Next‐Generation Material‐Based Sensors

Advanced Science, EarlyView.
This Review presents a structural framework that classifies neuromorphic sensing into near‐sensor and in‐sensor architectures, clarifying physical coupling between sensing and computation. The framework connects neural and synaptic device functions with recent advances in optical, mechanical, and chemical sensing, compares energy consumption and ...
Su Yeon Jung, Gwang Ya Kim, Sejin Kim, Hyunjong Seo, Se Gi Lee, Sang Min Won, Dongjoon Rhee, Joohoon Kang   +7 more
wiley   +1 more source

Hybrid Task- and Data-Parallelization on Heterogeneous Platform Using Model-Based Tool and Library Function Generation

IEEE Access
Modern embedded systems such as autonomous vehicles and robotics increasingly rely on high-performance computing to satisfy real-time and data-intensive demands.
Shanwen Wu, Qi Li, Masato Edahiro
doaj   +1 more source

Exploring Various Levels of Parallelism in High-Performance CRC Algorithms

IEEE Access, 2019
Modern processors have increased the capabilities of instruction-level parallelism (ILP) and thread-level parallelism (TLP). These resources, however, typically exhibit poor utilization on conventional cyclic redundancy check (CRC) algorithms.
Mucong Chi, Dazhong He, Jun Liu
doaj   +1 more source

Photonic‐Enabled Energy‐Efficient Transparent Neuromorphic Computing Devices: A Review

Advanced Science, EarlyView.
Transparent photonic neuromorphic computing devices merge optics and brain‐inspired computing to overcome von Neumann bottlenecks with ultrafast, low‐energy processing. By exploiting transparent oxides, 2D materials, phase‐change materials, and hybrid heterostructures, these platforms enable photonic synapses, memory, and logic for see‐through edge ...
Shuvaraj Ghosh, Ki‐bum Lee, Junghyeon Lee, Seunghee Cho, Malkeshkumar Patel, Hangfei Li, Yu Wen, Ye Zhou, Joondong Kim   +8 more
wiley   +1 more source

THE PERFORMANCE OF CONVOLUTIONAL NEURAL NETWORKS USING AN ACCELERATOR

Компютерні системи та інформаційні технології
The effectiveness of convolutional neural networks (CNNs) has been demonstrated across various fields, including computer vision, natural language processing, medical imaging, and autonomous systems.
Tymur ISAIEV, Tetiana KYSIL
doaj   +1 more source

Dual‐Mode Nanoporous SiO2 Memristors with Coexisting Volatile and Nonvolatile Dynamics for Reservoir Computing

Advanced Science, EarlyView.
A nanoporous SiO2 memristor enabling reconfigurable volatile and non‐volatile switching within a single device is demonstrated. The dual‐mode functionality supports both physical reservoir dynamics and synaptic weight storage, allowing unified hardware implementation of reservoir computing for temporal information processing, including image and ...
Bohao Ding, Tongjun Zhang, Li Shao, Nikolay Zhelev, Andrew L. Hector, Ruomeng Huang   +5 more
wiley   +1 more source