检索结果-内蒙古大学图书馆

4th International Conference on Electronic Information Engineering and Computer Science, EIECS 2024

作者： Xie, Yufei Wang, Wenlin Wu, Guohua Yu, Yang An, Ping Sun, Zibin Zhang, Haichuan Luo, Shengfeng Li, Yue School of Automation Wuhan University of Technology Wuhan China Sino-German College of Intelligent Manufacturing Shenzhen Technology University Shenzhen China Nuclear Power Institute of China Chengdu China National University of Defense Technology National Key Laboratory of Parallel and Distributed Computing Changsha China

ISBN: (纸本)9798331531409

In the field of nuclear energy, the Loss of Coolant Accident (LOCA) is recognized as one of the most severe types of nuclear reactor accidents, characterized by its complex physical processes and potentially catastrophic consequences. These challenges impose stringent requirements on safety analysis and emergency response. Accurate prediction and analysis of fluid behavior within pipelines under LOCA conditions are critical for evaluating accident outcomes and formulating response strategies. This paper introduces an innovative intelligent computing approach - a Physics-Informed Neural Networks (PINNs) model driven by both physical data and simulation data, specifically tailored for LOCA conditions. To address the challenges posed by multivariable and complex physical relationships, the six-equation two-fluid model is first simplified to represent the physical processes. Subsequently, a dual-driven PINNs network is developed, integrating both simulation data and physical constraints. The proposed model demonstrates a root mean square error (RMSE) of 0.02, a mean absolute error (MAE) of 0.044, and an R2 of 0.81 in predicting the outcomes under the six-equation model. © 2024 IEEE.

关键词： Nuclear energy

来源：评论

学校读者我要写书评

暂无评论

CHIME: A Cache-Efficient and High-Performance Hybrid Index on Disaggregated Memory 24

CHIME: A Cache-Efficient and High-Performance Hybrid Index o...

引用

30th ACM Symposium on Operating Systems Principles, SOSP 2024

作者： Luo, Xuchuan Shen, Jiacheng Zuo, Pengfei Wang, Xin Lyu, Michael R. Zhou, Yangfan National Key Laboratory of Parallel and Distributed Computing Changsha China School of Computer Science Fudan University Shanghai China Duke Kunshan University Kunshan China Huawei Cloud Shenzhen China Shanghai Key Laboratory of Intelligent Information Processing Shanghai China The Chinese University of Hong Kong Hong Kong Hong Kong

ISBN: (纸本)9798400712517

Disaggregated memory (DM) is a widely discussed datacenter architecture in academia and industry. It decouples computing and memory resources from monolithic servers into two network-connected resource pools. Range indexes are widely adopted by storage systems on DM to efficiently locate and query remote data. However, existing range indexes on DM suffer from either high computing-side cache consumption or high memory-side read amplifications. In this paper, we propose CHIME, a hybrid index combining B+ trees with hopscotch hashing, to achieve low cache consumption and low read amplifications simultaneously. There are three challenges in constructing CHIME on DM, i.e., the complicated optimistic synchronization, the extra metadata access, and the read amplifications introduced by hopscotch hashing. CHIME leverages 1) a three-level optimistic synchronization scheme to synchronize read and write operations with various granularities, 2) an access-aggregated metadata management technique to eliminate extra metadata accesses by piggybacking and replicating metadata, and 3) an effective hotness-aware speculative read mechanism to mitigate the read amplifications of hopscotch hashing. Experimental results show that CHIME outperforms the state-of-the-art range indexes on DM by up to 5.1× with the same cache size and achieves similar performance with up to 8.7× lower cache consumption. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

Efficient distributed parallel Aligning Reads and Reference Genome with Many Repetitive Subsequences Using Compact de Bruijn Graph

Efficient Distributed Parallel Aligning Reads and Reference ...

引用

International Symposium on parallel Architectures, Algorithms and Programming (PAAP)

作者： Yao Li Cheng Zhong Danyang Chen Jinxiong Zhang Mengxiao Yin Key Laboratory of Parallel School of Computer Electronics and Information Distributed Computing Technology in Guangxi Universities Guangxi University Nanning China

ISBN: (纸本)9781665496407

A large number of reads generated by the next generation sequencing platform will contain many repetitive subsequences. Effective localizing and identifying genomic regions containing repetitive subsequences will contribute to the subsequent genomic data analysis. To accelerate the alignment between large-scale short reads and reference genome with many repetitive subsequences, this paper develops a compact de Bruijn graph based short-read alignment algorithm on distributed parallel computing platform. The algorithm uses resilient distributed data sets (RDDS) to perform calculations in memory, and executes the broadcast method to distribute short reads and reference genome to the computing nodes to reduce the data communication time on the cluster system, and the number of RDD partitions is set to optimize the performance of parallel aligning algorithm. Experimental results on real datasets show that compared with the compact de Bruijn graph based sequential short-read alignment algorithm, our implemented distributed parallel alignment algorithm achieves good acceleration on the premise of obtaining the same correct alignment percentage as a whole, and compared with existing distributed parallel alignment algorithms, the implemented parallel algorithm can more quickly complete the alignment between large-scale short reads and reference genome with highly repetitive subsequences.

关键词： Sequential analysis Genomics Clustering algorithms distributed databases Programming Partitioning algorithms parallel architectures

来源：评论

学校读者我要写书评

暂无评论

Local-Adaptive Transformer for Multivariate Time Series Anomaly Detection and Diagnosis

Local-Adaptive Transformer for Multivariate Time Series Anom...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Xiaohui Zhou Yijie Wang Hongzuo Xu Mingyu Liu Ruyi Zhang National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Changsha China

Time series data are pervasive in varied real-world applications, and accurately identifying anomalies in time series is of great importance. Many current methods are insufficient to model long-term dependence, whereas some anomalies can be only identified through long temporal contextual information. This may finally lead to disastrous outcomes due to false negatives of these anomalies. Prior arts employ Transformers (i.e., a neural network architecture that has powerful capability in modeling long-term dependence and global association) to alleviate this problem; however, Transformers are insensitive in sensing local context, which may neglect subtle anomalies. Therefore, in this paper, we propose a local-adaptive Transformer based on cross-correlation for time series anomaly detection, which unifies both global and local information to capture comprehensive time series patterns. Specifically, we devise a cross-correlation mechanism by employing causal convolution to adaptively capture local pattern variation, offering diverse local information into the long-term temporal learning process. Furthermore, a novel optimization objective is utilized to jointly optimize reconstruction of the entire time series and matrix derived from cross-correlation mechanism, which prevents the cross-correlation from becoming trivial in the training phase. The generated cross-correlation matrix reveals underlying interactions between dimensions of multivariate time series, which provides valuable insights into anomaly diagnosis. Extensive experiments on six real-world datasets demonstrate that our model outperforms state-of-the-art competing methods and achieves 6.8%-27.5% $F_{1}$ score improvement. Our method also has good anomaly interpretability and is effective for anomaly diagnosis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Rethinking the distributed DNN Training Cluster Design from the Cost-effectiveness View

Rethinking the Distributed DNN Training Cluster Design from ...

引用

IEEE International Conference on High Performance computing and Communications (HPCC)

作者： Zhiquan Lai Yujie Liu Wei Wang Yanqi Hao Dongsheng Li National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Changsha China

As deep learning grows rapidly, model training heavily relies on parallel methods and there exist numerous cluster configurations. However, current preferences for parallel training focus on data centers, overlooking the financial constraints faced by most researchers. To attain the best performance within the cost limitation, we introduce a throughput-cost metric to accurately characterize clusters' cost-effectiveness. Based on this metric, we design a cost-effective cluster featuring the 3090 with NVLink. The experiment results demonstrate that our cluster achieves remarkable cost-effectiveness in various distributed model training schemes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Communication Analysis for Multidimensional parallel Training of Large-scale DNN Models

Communication Analysis for Multidimensional Parallel Trainin...

引用

IEEE International Conference on High Performance computing and Communications (HPCC)

作者： Zhiquan Lai Yanqi Hao Shengwei Li Dongsheng Li National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Changsha China

Multidimensional parallel training has been widely applied to train large-scale deep learning models like GPT-3. The efficiency of parameter communication among training devices/processes is often the performance bottleneck of large model training. Analysis of parameter communication mode and traffic has important reference significance for the research of interconnection network design and computing task scheduling to improve the training performance. In this paper, we analyze the parametric communication modes in typical 3D parallel training (data parallelism, pipeline parallelism, and tensor parallelism), and model the traffic in different communication modes. Finally, taking GPT-3 as an example, we present the communication in its 3D parallel training.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Area-NeRF: Area-based Neural Radiance Fields

Area-NeRF: Area-based Neural Radiance Fields

引用

Image Processing, Computer Vision and Machine Learning (ICICML), International Conference on

作者： Zonxin Ye Wenyu Li Peng Qiao Yong Dou National Key Laboratory of Parallel and Distributed Computing School of Computer National University of Defense Technology Changsha China

Neural Radiance Field (NeRF) has received widespread attention for its photo-realistic novel view synthesis quality. Current methods mainly represent the scene based on point sampling of ray casting, ignoring the influence of the observed area changing with distance. In addition, The current sampling strategies are all focused on the distribution of sampling points on the ray, without paying attention to the sampling of the ray. We found that the current ray sampling strategy for scenes with the camera moving forward severely reduces the convergence speed. In this work, we extend the point representation to area representation by using relative positional encoding, and propose a ray sampling strategy that is suitable for camera trajectory moving forward. We validated the effectiveness of our method on multiple public datasets.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Large Models Fine-tuning on Commodity Servers via Memory-balanced Pipeline parallelism

Efficient Large Models Fine-tuning on Commodity Servers via ...

引用

IEEE International Conference on High Performance computing and Communications (HPCC)

作者： Yujie Liu Zhiquan Lai Weijie Liu Wei Wang Dongsheng Li National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Changsha China

Large models have achieved impressive performance in many downstream tasks. Using pipeline parallelism to fine-tune large models on commodity GPU servers is an important way to make the excellent performance of large models available to the general public. Previous solutions fail to achieve an efficient memory-balanced pipeline parallelism. In this poster, we introduce a memory load-balanced pipeline parallel solution. This solution balances memory consumption across stages on commodity GPU servers via NVLink bridges. It establishes a new pathway to offload data from GPU to CPU by using the PCIe link of adjacent GPUs connected by the NVLink bridge. Furthermore, our method orchestrates offload operations to minimize the offload latency during large model fine-tuning. Experiments demonstrate that our solution can balance the memory footprint among pipeline stages without sacrificing training performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Highly parallelized Reinforcement Learning Training with Relaxed Assignment Dependencies

arXiv

引用

arXiv 2025年

作者： He, Zhouyu Qiao, Peng Li, Rongchun Dou, Yong Tan, Yusong College of Computer Science and Technology National University of Defense Technology China National Key Laboratory of Parallel and Distributed Computing National University of Defense Technology China

As the demands for superior agents grow, the training complexity of Deep Reinforcement Learning (DRL) becomes higher. Thus, accelerating training of DRL has become a major research focus. Dividing the DRL training process into subtasks and using parallel computation can effectively reduce training costs. However, current DRL training systems lack sufficient parallelization due to data assignment between subtask components. This assignment issue has been ignored, but addressing it can further boost training efficiency. Therefore, we propose a high-throughput distributed RL training system called TianJi. It relaxes assignment dependencies between subtask components and enables event-driven asynchronous communication. Meanwhile, TianJi maintains clear boundaries between subtask components. To address convergence uncertainty from relaxed assignment dependencies, TianJi proposes a distributed strategy based on the balance of sample production and consumption. The strategy controls the staleness of samples to correct their quality, ensuring convergence. We conducted extensive experiments. TianJi achieves a convergence time acceleration ratio of up to 4.37 compared to related comparison systems. When scaled to eight computational nodes, TianJi shows a convergence time speedup of 1.6 and a throughput speedup of 7.13 relative to XingTian, demonstrating its capability to accelerate training and scalability. In data transmission efficiency experiments, TianJi significantly outperforms other systems, approaching hardware limits. TianJi also shows effectiveness in on-policy algorithms, achieving convergence time acceleration ratios of 4.36 and 2.95 compared to RLlib and XingTian. TianJi is accessible at https://***/HiPRL/***. © 2025, CC BY.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Hemo-FS-SAM2: Few-Shot Tuning on Hemorrhage Segmentation of Segment Anything Model 2

Hemo-FS-SAM2: Few-Shot Tuning on Hemorrhage Segmentation of ...

引用

Advanced Algorithms and Control Engineering (ICAACE), International Conference on

作者： Xi Wang Peng Qiao Yong Dou National Key Laboratory of Parallel and Distributed Computing College of Computer Science and Technology National University of Defense Technology National University of Defense Technology Changsha China

ISBN: (数字)9798331535087

ISBN: (纸本)9798331535094

Surgical hemorrhage is a common occurrence in surgeries. Accurate segmentation of hemorrhage regions is important for surgical navigation and post-operative assessment. Some segmentation models focused on medical images, but its performance on hemorrhage data is limited. Besides, there is an essential challenge to annotate a large number of hemorrhage data, and previous segmentation methods struggle with complex hemorrhage characteristics like unclear boundaries and scattered targets. The Segment Anything Model 2 (SAM2) shows an significant zero-shot ability in general image segmentation, and is generalizable to fine-tuned on downstream tasks. However, it often faces limitations in hemorrhage segmentation task that lacking annotations. This paper proposes a fine-tuning approach for SAM2, significantly improving its performance on hemorrhage segmentation tasks with limited data. Our method provides better segmentation performance on few-shot hemorrhage data than SAM and SAM2 based models.

关键词： Image segmentation Adaptation models Accuracy Annotations Shape Surgery Data models Hemorrhaging Tuning Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：