检索结果-内蒙古大学图书馆

International Symposium on parallel and distributed Processing with Applications, ISPA

作者： Xiaohan Zuo Chunhua Xiao Qinglin Wang Chen Shi College of Computer Science Chongqing University China Key Laboratory of Dependable Service Computing in Cyber Physical Society Ministry of Education China National Key Laboratory of Parallel and Distributed Computing National University of Defense Technology Changsha China Laboratory of Digitizing Software for Frontier Equipment National University of Defense Technology Changsha China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

General Matrix Multiplication (GEMM) is a critical computational operation in scientific computing and machine learning domains. While traditional GEMM performs well on large matrices, it is inefficient in terms of data transfer and computation for small matrices. Many High-Performance Computing (HPC) tasks can be decomposed into large batches of small matrix multiplication operations. Multi-core Digital Signal Processors (DSPs) are commonly used to accelerate high-performance computing. We present a design for batched fusion small matrix multiplication (BFMM) tailored for multi-core DSP architecture. To address the inefficiencies and redundancy in storage and computational operations associated with batch small matrix multiplications, we designed several strategies. We design a matrix fusion concatenation strategy, an access coordination mechanism, and a mechanism for fragment aggregation. BFMM supports an efficient K-dimension multi-core parallelization strategy. The parameter constraint model makes BFMM highly portable. BFMM also includes a performance evaluation model that facilitates assessment and verification. Experimental results demonstrate that, compared to traditional GEMM (TGEMM) on multi-core DSP and traditional GEMM with concatenated data access (TGEMM Op), BFMM exhibits superior performance. For large batches of small matrices, our design achieves 1.21x to 18x higher performance than TGEMM Op on single-core DSP, while on multi-core DSP, it outperforms TGEMM Op by 1.14x to 18.1x.

关键词： Performance evaluation Scientific computing High performance computing Digital signal processors Redundancy Batch production systems Signal processing algorithms Computer architecture Machine learning Matrix decomposition

来源：评论

学校读者我要写书评

暂无评论

Big Data-aided Study of the Physical Process of Volume Ignition 6

Big Data-aided Study of the Physical Process of Volume Ignit...

引用

2022 6th International Conference on Mechanics, Mathematics and Applied Physics, ICMMAP 2022

作者： Liu, Pan Zhou, Yinzhao Xiong, Wenbin Liu, Chencheng Liu, Jie Liu, Gongjie Tang, Zhangchun Gao, Qiang Tianfu Innovation Energy Establishment Chengdu610000 China Nuclear and Radiation Safety Center MEE Beijing100000 China Central South University of Forestry and Technology Changsha410000 China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410073 China Beijng PARATERA Technology Corp. Ltd Beijing100000 China University of Electronic Science and Technology of China Chengdu610000 China

Volume ignition is a method of igniting a fuel as a whole by simultaneously achieving ignition conditions throughout the fuel zone. The basic criterion for ignition is that the thermonuclear energy is greater than the energy leakage at the fuel boundary, resulting in self-sustaining heating and deep combustion. Deuterium-tritium fuels are wrapped in medium to high Z media to reduce radiative leakage and achieve lower-temperature holistic ignition and non-equilibrium combustion, ultimately allowing the fuel to achieve high combustion efficiency. Volume ignition is the use of energy balance relations under the local thermodynamic equilibrium approximation to establish the energy balance equation for thermonuclear systems, and the system ignition threshold is obtained by solving this equation. By understanding the physical process, we believe that the non-equilibrium process is universal to the volume ignition process. Changes in external factors (density, boundaries, albedo, etc.) at the moment of ignition can have a significant impact on the development direction of the system, and an ignition system with a large surface density can nevertheless withstand a large amount of reverse work and continue to burn. The design of the ignition target tries to avoid these factors through margin design, but conversely, the rational use of these laws can further improve the design margin of the capsule. With the aid of big data, the volume ignition method is easier to calculate and has a shorter iteration time. The traditional way is to propose a model, set the material, and then perform the calculation while using big data can set any model and material for calculation. In this paper, a simple comparison will be made to find out that the efficiency of the physical design of a volume ignition target will be effectively improved with the aid of big data. Volume ignition targets can be used not only in Z-pinch-driven systems but also for laser-driven volume ignition. © Published un

关键词： Ignition

来源：评论

学校读者我要写书评

暂无评论

Optimizing Irregular-Shaped Matrix-Matrix Multiplication on Multi-Core DSPs

arXiv

引用

arXiv 2022年

作者： Yin, Shangfei Wang, Qinglin Hao, Ruochen Zhou, Tianyang Mei, Songzhu Liu, Jie Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410073 China College of Computer National University of Defense Technology Changsha410073 China

General Matrix Multiplication (GEMM) has a wide range of applications in scientific simulation and artificial intelligence. Although traditional libraries can achieve high performance on large regular-shaped GEMMs, they often behave not well on irregular-shaped GEMMs, which are often found in new algorithms and applications of high-performance computing (HPC). Due to energy efficiency constraints, low-power multi-core digital signal processors (DSPs) have become an alternative architecture in HPC systems. Targeting multi-core DSPs in FTm7032, a prototype CPU-DSPs heterogeneous processor for HPC, an efficient implementation - ftIMM - for three types of irregular-shaped GEMMs is proposed. FtIMM supports automatic generation of assembly micro-kernels, two parallelization strategies, and auto-tuning of block sizes and parallelization strategies. The experiments show that ftIMM can get better performance than the traditional GEMM implementations on multi-core DSPs in FT-m7032, yielding on up to 7.2× performance improvement, when performing on irregular-shaped GEMMs. And ftIMM on multi-core DSPs can also far outperform the open source library on multi-core CPUs in FT-m7032, delivering up to 3.1× higher efficiency. Copyright © 2022, The Authors. All rights reserved.

关键词： Digital signal processors

来源：评论

学校读者我要写书评

暂无评论

Modeling full information with graph network for joint entity-relation extraction 4

Modeling full information with graph network for joint entit...

引用

4th International Conference on Algorithms, Computing and Artificial Intelligence, ACAI 2021

作者： Wan, Qian Wei, Luona Chen, Xinhai Liu, Jie Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Hunan Changsha China College of Systems Engineering National University of Defense Technology Hunan Changsha China Laboratory of Software Engineering for Complex Systems National University of Defense Technology Hunan Changsha China

ISBN: (纸本)9781450385053

Fully capturing contextual information and analyzing the association between entity semantics and type is helpful for joint extraction task: 1) The context can reflect the part of speech and semantics of entity. 2) The entity type is closely related to the relation between entities. Previous research used to simply embed the contextual information into shallow layer of the model, ignoring the association between entity semantics and type. In this paper, we propose a graph network with full-information modeling to explicitly model different-level information in the text. The contextual information of entity is dynamically embedded in each span representation to improve the reasoning ability. To capture the fine-grained association between the semantics and type of entity, the graph network uses the feature of entity types to generate edge information between different nodes. Experimental results show that our model outperforms previous models on the CoNLL04 dataset and obtains competitive results on the SciERC dataset in both entity recognition and relation extraction. Extensive additional experiments further verify the effectiveness of the model. © 2021 ACM.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Optimizing All-to-All Collective Communication on Tianhe Supercomputer

Optimizing All-to-All Collective Communication on Tianhe Sup...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Jintao Peng Jie Liu Yi Dai Min Xie Chunye Gong National University of Defense Technology ChangSha China Science and Technology on Parallel and Distributed Processing Laboratory NUDT Changsha China Changsha and Laboratory of Software Engineering for Complex System NUDT Changsha China

All-to-all communication has a wide range of ap-plications in parallel applications like FFT. On most supercom-puters, there are multiple cores in a node. Message aggregation is an efficient method for smaller messages. Using multi-leader to aggregate message show significant improvement in intra-node overhead. However, compared to one-leader aggregation, existing multi-leader design incurs more message count and less aggregated message size. This paper proposes an Overlapped Multi-worker Multi-port all-to-all (OVALL) to scale the message size and parallelism of the aggregation algorithm. The algorithm exploits the all-to-all multi-core parallelism, concurrent commu-nication, and overlapping capabilities. Experiment results show that, compared to systems built-in MPI, OVALL'implementation achieves up to 5.9x or 18x speedup compared to MPI on different HPC systems. For the Fast Fourier Transform (FFT) application, OVALL is up to 2.7x (8192 cores, system A) or 5.6x (4800 cores, system B) faster compared to built-in MPI on peak performance.

关键词： Multicore processing Fast Fourier transforms Scalability Aggregates Supercomputers Libraries Pipeline processing

来源：评论

学校读者我要写书评

暂无评论

Big data collaborative artificial intelligence and high-performance computing to drive physical design of fusion

Big data collaborative artificial intelligence and high-perf...

引用

2022 International Conference on Cloud Computing, Big Data Applications and Software Engineering, CBASE 2022

作者： Liu, Pan Liu, Gongjie Liu, Jie Liu, Chencheng Xiong, Wenbin Tang, Zhangcun Zhou, Yinzhao Wang, Yonggang Yuan, Zhouming Beijng Paratera Technology Corp. Ltd Beijing100000 China National University of Defense Technology Science and Technology on Parallel and Distributed Processing Laboratory Changsha410073 China National University of Defense Technology Laboratory of Software Engineering for Complex Systems Changsha410073 China Central South University of Forestry and Technology Changsha410000 China Nuclear and Radiation Safety Center Mee Beijing100000 China University of Electronic Science and Technology of China Chengdu610000 China Tianfu Innovation Energy Establishment Chengdu610000 China

ISBN: (纸本)9798350396393

Controlled thermonuclear fusion has always been a dream pursued by mankind. However, the physical processes of controlled thermonuclear fusion are complex, requiring numerical simulations with high performance computing, and the amount of data generated by the physical processes on spatial, temporal and temperature scales is too large to be captured, managed, processed and collated in a reasonable time frame by mainstream software tools to achieve more aggressive fusion physical design. The data are too large to be captured, managed, processed, and collated into more aggressive targets for fusion physical design in a reasonable time by mainstream software tools. At the same time, the failure of fusion ignition can be caused by the distortion of various key physical quantities, and only by decomposing the process step by step and clarifying the changes of key physical quantities in the fusion physics process, can an effective mechanism be formed to prevent the distortion of key physical quantities from causing ignition failure in experimental physics. Big data in collaboration with artificial intelligence and high performance computing to drive the physical design of fusion is a novel avenue. By data acquisition with and pre-processing, this allows the creation of small sample libraries and deep learning. With supervised learning function convergence, incorporating solid/fluid computational methods, further network layering and cell expansion, the parameters of the physical model conforming to Lawson's criterion will become experimental physical parameters, and we find the new approach in Data capacity, Model combination approach, Type of material, Calculation speed, Optimisation of design iteration times, etc. are superior to the traditional approach. © 2022 IEEE.

关键词： Big data

来源：评论

学校读者我要写书评

暂无评论

OHTMA:an optimized heuristic topology-aware mapping algorithm on the Tianhe-3 exascale supercomputer prototype

引用

Frontiers of Information technology & Electronic Engineering 2020年第6期21卷 939-949页

作者： Yi-shui LI Xin-hai CHEN Jie LIU Bo YANG Chun-ye GONG Xin-biao GAN Sheng-guo LI Han XU Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense TechnologyChangsha 410073China

With the rapid increase of the size of applications and the complexity of the supercomputer architecture,topology-aware process mapping becomes increasingly *** communication cost has become a dominant constraint of the performance of applications running on the *** avoid a bad mapping strategy which can lead to terrible communication performance,we propose an optimized heuristic topology-aware mapping algorithm(OHTMA).The algorithm attempts to minimize the hop-byte metric that we use to measure the mapping *** incorporates a new greedy heuristic method and pair-exchange-based *** reduces the number of long-distance communications and effectively enhances the locality of the *** results on the Tianhe-3 exascale supercomputer prototype indicate that OHTMA can significantly reduce the communication costs.

关键词： High-performance computing Topology mapping Heuristic algorithm

来源：评论

学校读者我要写书评

暂无评论

Effective Anomaly Detection Based on Reinforcement Learning in Network Traffic Data

Effective Anomaly Detection Based on Reinforcement Learning ...

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Zhongyang Wang Yijie Wang Hongzuo Xu Yongjun Wang Science and Technology on Parallel and Distributed Processing Laboratory College of Computer National University of Defense Technology

Mixed-type data with both categorical and numerical features are ubiquitous in network security, but the existing methods are minimal to deal with them. Existing methods usually process mixed-type data through feature conversion, whereas their performance is downgraded by information loss and noise caused by the transformation. Meanwhile, existing methods usually superimpose domain knowledge and machine learning in which fixed thresholds are used. It cannot dynamically adjust the anomaly threshold to the actual scenario, resulting in inaccurate anomalies obtained, which results in poor performance. To address these issues, this paper proposes a novel Anomaly Detection method based on Reinforcement Learning, termed ADRL, which uses reinforcement learning to dynamically search for thresholds and accurately obtain anomaly candidate sets, fusing domain knowledge and machine learning fully and promoting each other. Specifically, ADRL uses prior domain knowledge to label known anomalies and uses entropy and deep autoencoder in the categorical and numerical feature spaces, respectively, to obtain anomaly scores combining with known anomaly information, which are integrated to get the overall anomaly scores via a dynamic integration strategy. To obtain accurate anomaly candidate sets, ADRL uses reinforcement learning to search for the best threshold. Detailedly, it initializes the anomaly threshold to get the initial anomaly candidate set and carries on the frequent rule mining to the anomaly candidate set to form the new knowledge. Then, ADRL uses the obtained knowledge to adjust the anomaly score and get the score modification rate. According to the modification rate, different threshold modification strategies are executed, and the best threshold, that is, the threshold under the maximum modification rate, is finally obtained, and the modified anomaly scores are obtained. The scores are used to re-carry out machine learning to improve the algorithm's accuracy for anomalo

关键词： Knowledge engineering Machine learning algorithms Conferences distributed databases Reinforcement learning Telecommunication traffic Network security

来源：评论

学校读者我要写书评

暂无评论

Entropy and Autoencoder-Based Outlier Detection in Mixed-Type Network Traffic Data

Entropy and Autoencoder-Based Outlier Detection in Mixed-Typ...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Zhongyang Wang Yijie Wang Zhenyu Huang Yongjun Wang Science and Technology on Parallel and Distributed Processing Laboratory College of Computer National University of Defense Technology

Mixed-type data containing categorical and numerical features are pervasive in real life, but very limited outlier detection methods are available for these data. Some existing methods handle mixed-type data by feature converting, whereas their performance is downgraded by information loss and noise caused by the transformation. Meanwhile, the existing general algorithms cannot combine the characteristics of outliers in specific fields, leading to an unsatisfying performance in actual scenarios, such as the field of network security. This paper proposes a novel Entropy and Autoencoder-based Outlier Detection in mixed-type network traffic data, termed EAOD, which combines characteristics of outliers in specific fields and machine learning models to detect outliers. EAOD utilizes the expert rules made by domain knowledge summarized based on characteristics of existing outliers to label known outlier data. It performs holoentropy and deep autoencoder for the category and numerical feature spaces, respectively, in unlabeled data to obtain outlier scores integrated to get the final outlier scores via a dynamic integration strategy. Especially in the numerical feature space, to fully mine known outlier behavior patterns, deep autoencoders of outlier and normal types are constructed separately to capture unknown outliers jointly. Experiments show that EAOD significantly outperforms eight state-of-the-art outlier detectors on seven real network traffic datasets.

关键词： Telecommunication traffic Machine learning Detectors Network security Feature extraction Entropy Data models

来源：评论

学校读者我要写书评

暂无评论

Attention-based Encoder-Decoder Recurrent Neural Networks for HTTP Payload Anomaly Detection

Attention-based Encoder-Decoder Recurrent Neural Networks fo...

引用

IEEE International Conference on Big Data and Cloud Computing (BdCloud)

作者： Shang Wu Yijie Wang Science and Technology on Parallel and Distributed Processing Laboratory College of Computer National University of Defense Technology

Attack payloads are often short segments hidden in HTTP requests; thus they can be found by HTTP payload anomaly detection. Deep learning models learn data features during training without manual feature extraction, and better performance has received more attention. Recurrent Neural Network models process sequences directly, which are widely used in payload anomaly detection. However, due to the gradient vanishing problem, RNN has limits on processing the long sequences. Meanwhile, RNN uses its final hidden state for detection and pays more attention to the content of the end of the payload. Besides, deep learning generally lacks interpretability. The paper proposes an unsupervised deep learning model for HTTP payload Anomaly Detection, namely Attention-based Encoder-Decoder Recurrent Neural Networks Anomaly Detection model (AEDRAD). AEDRAD utilizes the encoder-decoder RNN and attention mechanism to detect anomalies by reconstructing the original sequences. AEDRAD filters the fields of HTTP protocol that cannot exist anomalies, focusing on the suspicious segments. Through the encoder-decoder network, the normal payload can be well-reconstructed while the anomaly payload fails. With the attention mechanism, AEDRAD generates practical features for further anomaly detection from a global perspective. Meanwhile, it marks abnormal fragments visually, which is conducive to a subsequent analysis by experts. The experimental results show that AEDRAD significantly outperforms three state-of-the-art unsupervised algorithms on two real datasets.

关键词： Deep learning Training Recurrent neural networks Protocols Focusing Manuals Feature extraction

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：