Results 51 to 60 of about 39,047 (336)

Simulee: Detecting CUDA Synchronization Bugs via Memory-Access Modeling

International Conference on Software Engineering, 2020
While CUDA has become a mainstream parallel computing platform and programming model for general-purpose GPU computing, how to effectively and efficiently detect CUDA synchronization bugs remains a challenging open problem.
Mingyuan Wu, Yicheng Ouyang, Husheng Zhou, Lingming Zhang, Cong Liu, Yuqun Zhang   +5 more
semanticscholar   +1 more source

Transistor‐Level Activation Functions via Two‐Gate Designs: From Analog Sigmoid and Gaussian Control to Real‐Time Hardware Demonstrations

Advanced Materials, EarlyView.
Screen gate‐based transistors are presented, enabling tunable analog sigmoid and Gaussian activations. The SA‐transistor improves MRI classification accuracy, while the GA‐transistor supports precise Gaussian kernel tuning for forecasting. Both functions are implemented in a single device, offering compact, energy‐efficient analog AI processing ...
Junhyung Cho, Youngmin Han, Won Woo Lee, Youngwoo Yoo, Kannan Udaya Mohanan, Chang‐Hyun Kim, Junhwan Choi, Young‐Joon Kim, Wonjun Shin, Hocheon Yoo   +9 more
wiley   +1 more source

RT-CUDA: A Software Tool for CUDA Code Restructuring [PDF]

International Journal of Parallel Programming, 2016
Recent development in graphic processing units (GPUs) has opened a new challenge in harnessing their computing power as a new general purpose computing paradigm. However, porting applications to CUDA remains a challenge to average programmers, which have to package code in separate functions, explicitly manage data transfers between the host and device
Khan, Ayaz H., Al-Mouhamed, Mayez, Al-Mulhem, Muhammed, Ahmed, Adel F.   +3 more
openaire   +2 more sources

ChicGrasp: Imitation‐Learning‐Based Customized Dual‐Jaw Gripper Control for Manipulation of Delicate, Irregular Bio‐Products

Advanced Robotics Research, EarlyView.
Automated poultry processing lines still rely on humans to lift slippery, easily bruised carcasses onto a shackle conveyor. Deformability, anatomical variance, and hygiene rules make conventional suction and scripted motions unreliable. We present ChicGrasp, an end‐to‐end hardware‐software co‐designed imitation learning framework, to offer a ...
Amirreza Davar, Zhengtong Xu, Siavash Mahmoudi, Pouya Sohrabipour, Chaitanya Pallerla, Yu She, Wan Shou, Philip Glen Crandall, Dongyi Wang   +8 more
wiley   +1 more source

Methods of consumer gpu virtualization for cloud services: comparative analysis of performance and latency

Вестник Самарского университета: Естественнонаучная серия
This paper investigates methods of consumer graphics processor (GPU) virtualization for cloud services applications. A comparative analysis of GPU passthrough, SR-IOV, MIG, and time-sliced vGPU technologies is conducted.
A. E. Bazhenov, D. S. Danilov, A. V. Pape, A. V. Tsibikin   +3 more
doaj   +1 more source

Automated Alignment Powered by Computer Vision Streamlines the Two‐Photon Polymerization‐Based Micro 3D Printing of Multiscale and Multimaterial Structures

Advanced Intelligent Discovery, EarlyView.
Two‐photon polymerization enables high‐resolution microfabrication, but performing alignment when printing multiple structures is difficult. Here, we present a fast, robust, and open‐source protocol for automated alignment on Nanoscribe systems. Achieving ≈0.4 μm accuracy in under 5 s, our protocol reduces time and error in multimaterial printing. This
Daniel Maher, Valentina Mendoza Martinez, Marcin Piotr Piekarczyk, Stefan Sützl, Ada‐Ioana Bunea   +4 more
wiley   +1 more source

Air Traffic Management Using a GPU-Accelerated Genetic Algorithm

Transport and Telecommunication, 2023
Air traffic management is becoming highly complex with the rapid increase in the number of commercial and cargo flights, leading to increased traffic congestion and flight delays.
Rampure Rahul, Tiruvallur Raghav, Acharya Vybhav, Navad Shashank, Preethi P   +4 more
doaj   +1 more source

Efficient Use of GPU Memory for Large-Scale Deep Learning Model Training

Applied Sciences, 2021
To achieve high accuracy when performing deep learning, it is necessary to use a large-scale training model. However, due to the limitations of GPU memory, it is difficult to train large-scale training models within a single GPU.
Hyeonseong Choi, Jaehwan Lee
doaj   +1 more source

A Fully Soft Sensing Suit With Optimal Sensor Placement for Real‐Time Motion Tracking

Advanced Intelligent Systems, EarlyView.
A fully soft, skin‐conformable sensing suit integrating stretchable sensors, liquid metal wiring, and soft electrodes was developed using direct ink writing, with sensor placement optimized through an automated algorithmic pipeline. This system enables accurate and unobtrusive real‐time motion tracking, providing a scalable, material‐based solution to ...
Jinhyeok Oh, Joonbum Bae
wiley   +1 more source

An Efficient Method for Defining Multivariate Functions Using Expression Templates for Arrays in C++ and CUDA [PDF]

مجله مدل سازی در مهندسی, 2018
In this paper an efficient method for defining multi-variable functions using expression templates for array computations in computational fluid dynamics simulations in C++ is introduced.
Hossein Mahmoodi Darian
doaj   +1 more source