Results 81 to 90 of about 186,522 (338)

Design of Network on Chip Routing Units Based on Virtual Conflict Array [PDF]

open access: yesJisuanji gongcheng, 2019
Network on Chip(NoC) routing units share the input buffers and only allow sequential access to data,which limits the speed and efficiency of communication on chip.To improve the parallelism of NoC,a routing units architecture based on virtual conflict ...
YANG Tianhao, SUN Jin
doaj   +1 more source

Data-Parallel String-Manipulating Programs [PDF]

open access: yesACM SIGPLAN Notices, 2015
String-manipulating programs are an important class of programs with applications in malware detection, graphics, input sanitization for Web security, and large-scale HTML processing. This paper extends prior work on BEK, an expressive domain-specific language for writing string-manipulating programs, with algorithmic insights that make BEK both ...
Veanes, M   +3 more
openaire   +2 more sources

Exploiting Data Parallelism in the yConvex Hypergraph Algorithm for Image Representation using GPGPUs

open access: yes, 2013
To define and identify a region-of-interest (ROI) in a digital image, the shape descriptor of the ROI has to be described in terms of its boundary characteristics. To address the generic issues of contour tracking, the yConvex Hypergraph (yCHG) model was
Agarwal, Tejaswi   +2 more
core   +1 more source

Real‐Time and High‐Resolution NIR‐II‐L Imaging of Netrin‐1‐Mediated Neurovascular Coupling Driven by Dose‐Dependent Electroacupuncture

open access: yesAdvanced Science, EarlyView.
Disrupted neurovascular coupling contributes to ovarian dysfunction in polycystic ovary syndrome. Using near‑infrared II long‑wavelength imaging, this study visualizes how electroacupuncture dynamically regulates ovarian blood vessels in vivo. Immediate and cumulative stimulation elicit distinct vascular responses via a Netrin‐1 dependent mechanism ...
Yicong Wang   +15 more
wiley   +1 more source

A snapshot of parallelism in distributed deep learning training

open access: yesRevista Colombiana de Computación
The accelerated development of applications related to artificial intelligence has generated the creation of increasingly complex neural network models with enormous amounts of parameters, currently reaching up to trillions of parameters.
Hairol Romero-Sandí   +2 more
doaj   +1 more source

Al Nanoparticle‐Decorated Metal Oxide Synaptic Transistors for Ultralow‐Energy Neuromorphic Computing with Wide Dynamic Range

open access: yesAdvanced Science, EarlyView.
A nanoparticle‐engineered electrolyte‐gated memtransistor is introduced as a materials‐level strategy to overcome the intrinsic trade‐off between energy consumption and synaptic precision. By embedding aluminum nanoparticles at the oxide–electrolyte interface to modulate ion trapping dynamics, the device achieves stable multistate plasticity under ...
Jun‐Gyu Choi   +4 more
wiley   +1 more source

Pipeline Parallelism With Elastic Averaging

open access: yesIEEE Access
To accelerate the training speed of massive DNN models on large-scale datasets, distributed training techniques, including data parallelism and model parallelism, have been extensively studied.
Bongwon Jang, In-Chul Yoo, Dongsuk Yook
doaj   +1 more source

A Logical Model and Data Placement Strategies for MEMS Storage Devices

open access: yes, 2008
MEMS storage devices are new non-volatile secondary storages that have outstanding advantages over magnetic disks. MEMS storage devices, however, are much different from magnetic disks in the structure and access characteristics.
Kim, Min-Soo   +3 more
core   +1 more source

Tunable Negative Thermal Expansion in Fe/Cr‐Substituted Nd2Co17 Compounds via Magnetoelastic Coupling

open access: yesAdvanced Science, EarlyView.
This study achieves anisotropic thermal expansion tuning in Nd2(Co1‐xFex)17‐yCry compounds via a magnetoelastic strategy. Variable‐temperature synchrotron X‐ray diffraction reveals that increased Fe content induces switchable lattice responses. Compositional control reduces the volume expansion coefficient αV by 20% (x═0.7) and modulates TC (442–625 K),
Jiayuan Li   +8 more
wiley   +1 more source

Data Parallelism in Java

open access: yes, 1998
Java supports threads and remote method invocation, but it does not support either data-parallel or distributed programming. This paper discusses Java’s shortcomings with respect to data-parallel programming and presents workarounds. The technical contributions of this paper are twofold: a source-to-source transformation that maps foral l statements ...
openaire   +2 more sources

Home - About - Disclaimer - Privacy