The particle counting part in the unified gas-kinetic wave–particle method on graphics processing unit (GPU) devices is computationally intensive. This paper introduces a piecewise-hierarchical (P-H) particle counting strategy tailored for the Single Instruction Multiple Threads architecture, which leverages GPU memory hierarchy to reduce access conflicts. The strategy was evaluated based on throughput, roofline performance, and computation time metrics. Compared to the global counting strategy, the P-H approach achieved a 3.37× speedup for the particle counting kernel, and the overall program experienced a performance boost of more than 30%. 

A piecewise-hierarchical particle count method suitable for the implementation of the unified gas-kinetic wave–particle method on graphics processing unit devices

The popularization of graphic processing units (GPUs) has led to their extensive use in high-performance numerical simulations. The Lattice Boltzmann Methodology (LBM) is a general framework for constructing efficient numerical fluid simulations. In this scheme, the fluid quantities are approximated on a structured grid. At each time step, a shift-relaxation process is applied, where each kinetic value is shifted to the corresponding direction in the lattice. Thanks to its simplicity, the LBM is subject to many software optimizations. State-of-the-art techniques aim at adapting the LBM scheme to improve the computational throughput on modern processors. Currently, most effort is put into optimizing this process on GPUs, as their architecture is highly suited for this type of computation. A bottleneck of GPU implementations is that the data size of the simulation is limited by the GPU memory. This restricts the number of volume elements and, therefore, the degree of precision one can obtain. In this work, we divide the lattice structure into multiple subsets that can be executed individually. This allows the work to be distributed among different processing units at the cost of increased complexity and memory transfers. But the constraint on GPU memory is relaxed, as the subsets can be made as small as needed. Additionally, we use the task-based approach for parallelizing the application, which allows the computation to be efficiently distributed among multiple processing units.

Parallelization of the Lattice-Boltzmann schemes using the task-based method

The algorithm for hologram mask design plays a crucial role in determining the final quality of the reconstructed pattern in diffraction lithography or its feasibility for fabricating large-scale integrated circuits. This paper presents an approach for generating hologram masks of large space-bandwidth products (SBP), effectively reducing the computation complexity and memory requirements without necessitating additional computer power. By leveraging the GS algorithm and incorporating area division and an overlapping constraint, this method facilitates the generation of masks of large SBP, which is beneficial for use in diffraction lithography. Moreover, it can be optimized for parallel operations to reduce computation time significantly, enabling efficient generalization of diffraction masks of large SBP.

An Optimized Iterative Method for Generating Hologram Mask of Large Space-bandwidth-product

High-Performance Computing (HPC) workloads generate large volumes of data at high-frequency during their execution, which needs to be captured concurrently at scale. These workloads exploit accelerators such as GPU for faster performance. However, the limited onboard high-bandwidth memory (HBM) on the GPU, and slow device-to-host memory PCIe interconnects lead to I/O overheads during application execution, thereby exacerbating their overall runtime. To overcome the aforementioned limitations, techniques such as compression and asynchronous transfers have been used by data management runtimes. However, compressing small blocks of data leads to a significant runtime penalty on the application. In this paper, we design and develop strategies to optimize the trade-off between compressing checkpoints instantly and enqueuing transfers immediately versus accumulating snapshots and delaying compression to achieve faster compression throughput. Our evaluations on synthetic and real-life workloads for different systems and workload configurations demonstrate 1.3 × to 8.3 × speedup compared to the existing checkpoint approaches.

Towards Efficient I/O Pipelines Using Accumulated Compression

An Optimized Iterative Method for Generating Diffractive Mask of Large Space-Bandwidth-Product

An out-of-core stencil computation code handles large data whose size is beyond the capacity of GPU memory. Whereas, such an code requires streaming data to and from the GPU frequently. As a result, data movement between the CPU and GPU usually limits the performance. In this work, compression-based optimizations are proposed. First, an on-the-ﬂy compression technique is applied to an out-of-core stencil code, reducing the CPU-GPU memory copy. Secondly, a single working buﬀer technique is used to reduce GPU memory consumption. Experimental results show that the stencil code using the proposed techniques achieved 1.1 × speed and reduced GPU memory consumption by 33.0% on an NVIDIA Tesla V100 GPU.

Compression-Based Optimizations for Out-of-Core GPU Stencil Computation

Studies on generating dance sequences from music can greatly promote dance teaching and animation production. However, the current results of such tasks are not very satisfactory. Due to the lack of consideration of human body structure, unreal phenomena such as slippery feet and twisted joints will appear in the resulting movements. Since only a single correlation feature between music and dance is learned, there is a problem of poor matching between generated actions and music. To solve these problems, we propose a learning framework based on GAN and a multi-feature fusion strategy to realize the mapping from music to dance. We use two discriminators to constrain style and authenticity respectively to make our actions generated by the network more coherent and natural, which is essential in producing authentic dance sequences. Moreover, we extracted the three characteristics of music style, beat, and structure. Then we used the feature fusion method to obtain a comprehensive feature representation, which guarantees the consistency of the generated dance sequence with the input music to the greatest extent. The experimental quantitative and qualitative results show that our method can generate accurate, consistent, beat-matching dance movements from music. In addition, we also collected and produced a data set containing music features and corresponding pose sequences, which is convenient for pose generation based on music features.

Music to Dance: Motion Generation Based on Multi-Feature Fusion Strategy

Convolutional neural networks have achieved great success in many fields. However, the practical application of convolutional neural networks is hindered due to the high consumption of memory and computational. In this paper, we propose a two-stage method for model compression and acceleration. More specifically, the training process mainly includes the search stage and train stage, the approach is abbreviated as ST. In the search stage, we first search and remove the unnecessary parts of a large pre-trained network (named supernet) by certain evaluation criteria to get a pruned network. Then the weights in the pruned network are initialized to get a small network (called a subnet). During the training stage, the supernet is untrainable, and the subnet will be trained under the supervision of the supernet. The knowledge extracted from the supernet will be transmitted to the subnet, then the subnet will be able to learn from the dataset and the knowledge at the same time. We have proved the effectiveness of our method through implement extensive experiments on several state-of-the-art CNN models (including VGGNet, ResNet, and DenseNet). The subnet only with 1/10 parameters and 1/2 calculations achieves more significant performance than the supernet.

Search-and-Train: Two-Stage Model Compression and Acceleration.

Recent the-state-of-the-art image-based three-dimensional (3D) reconstruction methods that represent 3D shapes mainly using triangular mesh because of its memory efficiency and ability to present surface detail of objects compared to voxel and point cloud. Previous works usually follow an encoding and decoding pattern. A deep neural network to extract the features from the picture and reconstruct the 3D structure. It is a typical supervised learning process, requiring loss function to supervise the training. No existing works directly calculate the loss between the reconstruction mesh and ground truth mesh. Instead, they indirectly used the Chamfer Distance (CD) between point clouds as the loss. Most of the previous works focus on the encoding and decoding parts instead of the loss and CD is used for all works. However, when CD is applied to two point clouds with the same number of points, some points can match any number of points in another point cloud, so some points will be less involved in calculating the loss function, which will reduce the utilization of information. Therefore, We propose a new point matching strategy to calculate the loss. The point matching strategy we proposed limits the maximum number of matches for each point, allowing more points to be more involved in the loss calculation, thereby improving the information utilization rate. Experiments on single view reconstruction (SVR) and auto-encoding methods show that this new loss method can replace CD in this type of works and has better training results and 3D reconstruction quality.

A Point Matching Strategy of 3D Loss Function for Single RGB Images Deep Mesh Reconstruction

A compression-based memory-efficient optimization for out-of-core GPU stencil computation

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Xin Deng

Author Tools

Chat about Author