Results 41 to 50 of about 175,847 (284)

DeLTA: GPU Performance Model for Deep Learning Applications with In-depth Memory System Traffic Analysis

, 2019
Training convolutional neural networks (CNNs) requires intense compute throughput and high memory bandwidth. Especially, convolution layers account for the majority of the execution time of CNN training, and GPUs are commonly used to accelerate these ...
Chatterjee, Niladrish, Erez, Mattan, Lee, Donghyuk, Lym, Sangkug, O'Connor, Mike   +4 more
core   +1 more source

Reconfigurable and High-Efficiency Password Recovery Algorithms Based on HRCA

IEEE Access, 2021
Cryptographic algorithms are widely used in information security fields such as network protocol authentication and commercial encryption software. Password recovery based on the hash algorithm is an important means of electronic forensics, encrypted ...
Bin Li, Feng Feng, Xiaojie Chen, Yan Cao
doaj   +1 more source

Gunrock: GPU Graph Analytics

, 2017
For large-scale graph analytics on the GPU, the irregularity of data access and control flow, and the complexity of programming GPUs, have presented two significant challenges to developing a programmable high-performance graph library.
Davidson, Andrew, Liu, Weitang, Osama, Muhammad, Owens, John D., Pan, Yuechao, Riffel, Andy T., Wang, Leyuan, Wang, Yangzihao, Wu, Yuduo, Yang, Carl, Yuan, Chenshan   +10 more
core   +2 more sources

WTF, GPU! computing twitter's who-to-follow on the GPU [PDF]

Proceedings of the second ACM conference on Online social networks, 2014
In this paper, we investigate the potential of GPUs for performing link structure analysis of social graphs. Specifically, we implement Twitter's WTF ("Who to Follow") recommendation system on a single GPU. Our implementation shows promising results on moderate-sized social graphs. It can return the top-K relevant users for a single user in 172 ms when
Geil, Afton, Wang, Yangzihao, Owens, John D   +2 more
openaire   +3 more sources

Structural instability impairs function of the UDP‐xylose synthase 1 Ile181Asn variant associated with short‐stature genetic syndrome in humans

FEBS Letters, EarlyView.
The Ile181Asn variant of human UDP‐xylose synthase (hUXS1), associated with a short‐stature genetic syndrome, has previously been reported as inactive. Our findings demonstrate that Ile181Asn‐hUXS1 retains catalytic activity similar to the wild‐type but exhibits reduced stability, a looser oligomeric state, and an increased tendency to precipitate ...
Tuo Li, Pedro A. Sánchez‐Murcia, Bernd Nidetzky   +2 more
wiley   +1 more source

vkpolybench: A crossplatform Vulkan Compute port of the PolyBench/GPU benchmark suite

SoftwareX, 2021
PolyBench is a well-known set of benchmarks characterized by embarrassingly parallel kernels able to run on Graphic Processing Units (GPUs). While Polybench GPU kernels leverage well-established GP-GPU APIs such as CUDA and OpenCL, in this paper we ...
Nicola Capodieci, Roberto Cavicchioli
doaj   +1 more source

Peroxidasin enables melanoma immune escape by inhibiting natural killer cell cytotoxicity

Molecular Oncology, EarlyView.
Peroxidasin (PXDN) is secreted by melanoma cells and binds the NK cell receptor NKG2D, thereby suppressing NK cell activation and cytotoxicity. PXDN depletion restores NKG2D signaling and enables effective NK cell–mediated melanoma killing. These findings identify PXDN as a previously unrecognized immune evasion factor and a potential target to improve
Hsu‐Min Sung, David Bickel, Lena C. M. Krause, Daria Ezeriņa, Christian Ickes, Julian Wojtachnia, Christine S. Gibhardt, Magdalena Shumanska, Khadija Wahni, Andrea Paluschkiwitz, Julia Malo Pueyo, Ekaterina Baranova, Wim Vranken, Hedwig Stanisz, Ioana Stejerean‐Todoran, Michael P. Schön, Joris Messens, Ivan Bogeski   +17 more
wiley   +1 more source

GPU-Accelerated Real-Time Stixel Computation [PDF]

2017 IEEE Winter Conference on Applications of Computer Vision (WACV), 2017
The Stixel World is a medium-level, compact representation of road scenes that abstracts millions of disparity pixels into hundreds or thousands of stixels. The goal of this work is to implement and evaluate a complete multi-stixel estimation pipeline on an embedded, energy-efficient, GPU-accelerated device.
Hernandez-Juarez, Daniel, Espinosa, Antonio, Vázquez, David, López, Antonio Manuel, Moure, Juan Carlos   +4 more
openaire   +2 more sources

Analysing the significance of small conformational changes and low occupancy states in serial crystallographic data

FEBS Open Bio, EarlyView.
This protocol paper outlines methods to establish the success of a time‐resolved serial crystallographic experiment, by means of statistical analysis of timepoint data in reciprocal space and models in real space. We show how to amplify the signal from excited states to visualise structural changes in successful experiments.
Jake Hill, Yelyzaveta Pulnova, Elke De Zitter, Helen M. Ginn, Briony A. Yorke   +4 more
wiley   +1 more source

Architecture-Aware Optimization on a 1600-core Graphics Processor [PDF]

, 2011
The graphics processing unit (GPU) continues to make significant strides as an accelerator in commodity cluster computing for high-performance computing (HPC).
Daga, Mayank, Feng, Wu-chun, Scogland, Thomas R.W.   +2 more
core   +1 more source