Results 131 to 140 of about 2,053,524 (303)

Eventify Meets Heterogeneity: Enabling Fine-Grained Task-Parallelism on GPUs

Many scientific computing algorithms barely provide sufficient data-parallelism to exploit the ever-increasing hardware parallelism of today’s heterogeneous computing environments.
Morgenstern, Laura
core   +1 more source

Exploitation of Task Level Parallelism [PDF]

, 2014
Existing many systems were supporting task level parallelism usually involving the process of task creation and synchronization. The synchronization of task requires the clear definition about existing dependencies in a program or data-flow restraints ...
Mangal, Mayank
core  

A mathematical model of knee joint line orientation: Independent effects of femoral and tibial coronal anatomy and limits on achieving a horizontal joint line

Knee Surgery, Sports Traumatology, Arthroscopy, EarlyView.
Abstract Purpose Natural knee joint line orientation (JLO) relative to the ground remains poorly understood. A JLO parallel to the floor is commonly assumed to favour balanced medial‐lateral compartment loading, but the independent influence of femoral and tibial coronal anatomy on JLO during bipodal and single‐limb stance has not been quantified. This
Gautier Beckers, Mathias Hengartner, Marin Lavigne, Marc‐Olivier Kiss, Vincent Massé, Pascal‐André Vendittoli   +5 more
wiley   +1 more source

Intrinsically Design‐Rule‐Compliant Nanophotonic Inverse Design via Learned Generative Manifolds

Laser &Photonics Reviews, EarlyView.
A generative reparameterization framework for nanophotonic inverse design is introduced, restricting optimization to a learned manifold of design‐rule‐compliant geometries. Unlike conventional penalty‐based approaches, fabrication constraints are encoded intrinsically within the design representation.
Bahrem Serhat Danis, Demet Baldan Desdemir, Enes Akcakoca, Zeynep Ipek Yanmaz, Gulzade Polat, Ahmet Onur Dasdemir, Aytug Aydogan, Abdullah Magden, Emir Salih Magden   +8 more
wiley   +1 more source

Efficient motion estimation and discrete cosine transform implementation using the graphics processing units.

PLoS ONE
Motion Estimation (ME) and the two-dimensional (2D) discrete cosine transform (2D-DCT) are both computationally expensive parts of HEVC standard, therefore real-time performance of the HEVC may not be free from glitches. To address this issue, this study
Shahrukh Agha, Farmanullah Jan, Haroon Ahmed Khan, Muhammad Kaleem, Mansoor Khan   +4 more
doaj   +1 more source

Energy-aware-task-parallelism for efficient dynamic voltage, and frequency scaling, in CGRAs

, 2013
Today, coarse grained reconfigurable architectures (CGRAs) host multiple applications, with arbitrary communication and computation patterns. Each application itself is composed of multiple tasks, spatially mapped to different parts of platform ...
Kolin Paul, Ahmed Hemani, Syed. M. A. H. Jafri, Jafri, Syed Mohammad Asad Hassan,, Hemani, Ahmed,, Tajammul, Muhammad Adeel,, Plosila, Juha,, Juha Plosila, Paul, Kolin,, Hannu Tenhunen, Muhammad Adeel Tajammul, Tenhunen, Hannu,   +11 more
core   +1 more source

Phase Multiplexed Optical Computing: Reconfiguring a Multi‐Task Diffractive Optical Processor Using Illumination Phase Diversity

Laser &Photonics Reviews, EarlyView.
A monochrome multi‐task diffractive optical network architecture is designed to leverage illumination‐phase multiplexing to dynamically reconfigure its output function and accurately implement a large set of complex‐valued linear transformations between an input and an output aperture.
Xiao Wang, Aydogan Ozcan
wiley   +1 more source

Parallel Vector Processing in Optical Computing With Jacobi Time‐Wave Packets

Laser &Photonics Reviews, EarlyView.
Functionality of a single neuron and state of the art photonic acceleration with a Mach‐Zehnder‐interferometer (MZI) mesh (i), a crossbar array (ii) and a micro ring resonator (MRR) array (iii). In the proposed Jacobi time‐wave packet approach (iv) a single modulator (here a ring modulator) is performing the whole scalar vector product in a single ...
Janosch Meier, Khaleda Mallick, Abhinand Venugopalan, Milind Bagul, Gouri Krishnan, Thomas Schneider   +5 more
wiley   +1 more source

Dual‐Metasurface‐Integrated Fiber Array for Bidirectional Cylindrical‐Vector‐Beam Multiplexed Communication Using Off‐Axis Spatial Manipulation

Laser &Photonics Reviews, EarlyView.
A dual‐layer metasurface is integrated on a fiber‐array facet to realize compact, plug‐and‐play free‐space optical links with cylindrical vector beam (CVB) space‐division multiplexing. An optimized off‐axis collimation layer steers multiple channels, while a Dammann vortex layer performs polarization–topology conversion for bidirectional CVB (de ...
Xipeng Lu, Jinke Li, Jin Tae Kim, Shaoyan Li, Hongliang Li, Duk‐Yong Choi, Sang‐Shin Lee   +6 more
wiley   +1 more source

Parallel Task Execution

, 2018
Puppet is a great tool for making changes on systems, and ensuring that those changes happen. But Puppet is not intended to make this happen on many systems at the same time. Puppet is intended for eventual compliance over time. Each agent checks in over a period of time, al- lowing the puppetmaster to process only a few at a time.
openaire   +2 more sources