检索结果-内蒙古大学图书馆

Evaluating matrix multiplication-based convolution algorithm on multi-core digital signal processors

Guofang Keji Daxue Xuebao/Journal of National University of Defense technology 2023年第1期45卷 86-94页

作者： Wang, Qinglin Pei, Xiangdong Liao, Linyu Wang, Haoxu Li, Rongchun Mei, Songzhu Li, Dongsheng College of Computer Science and Technology National University of Defense Technology Changsha410073 China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410073 China

The matrix multiplication-based convolutional algorithm, which can efficiently implement convolutions with different parameters, is the first choice of convolution performance optimization for a given chip. Based on the architecture of Phytium heterogeneous multi-core DSPs(digital signal processors) developed by National University of Defense technology and the characteristic of the matrix multiplication-based convolutional algorithm, a parallel implementation of the matrix multiplication-based convolutional algorithm (called ftmEConv) for different convolutions on multi-core DSPs was proposed. The ftmEConv consists of four parallelized parts(input feature maps transformation, filter transformation, matrix multiplication, and output feature maps transformation), all of which were optimized for multi-core DSPs, and the performance of each part was improved by effectively exploiting the potential of all functional units in DSP cores. The experimental results demonstrate that ftmEConv achieves computational efficiency of up to 42.90%. Compared with other implementations of the matrix multiplication-based convolutional algorithm on heterogeneous chips, ftmEConv gets a speedup of up to 7.79 times. © 2023 National University of Defense technology. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A neuro-genetic approach for inferring gene regulatory networks from gene expression data 22

A neuro-genetic approach for inferring gene regulatory netwo...

引用

Proceedings of the 9th International Conference on Bioinformatics Research and Applications

作者： Guo Mao Zhengbin Pang Jie Liu Ke Zuo Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology China

ISBN: (纸本)9781450396868

Accurate prediction of gene regulation rules is important for understanding complex life processes. Existing computational algorithms designed for bulk transcriptome datasets typically require a large number of time points to infer gene regulatory networks (GRNs), are suitable for a small number of genes, and cannot efficiently detect potential regulatory relationships. We propose an approach based on a deep learning framework to reconstruct GRNs from bulk transcriptome datasets, assuming that the expression levels of transcription factors involved in gene regulation are strong predictors of the expression of their target genes. The algorithm uses multilayer perceptrons to infer the regulatory relationship between multiple transcription factors and a gene, and uses genetic algorithms to search for the best regulatory gene combination. The results show that our approach is more accurate than other methods for reconstructing gene regulatory networks on real-world and simulated bulk transcriptome gene expression datasets.

关键词： gene regulatory networks datasets multilayer perceptrons genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Location-Aware Multi-user Resource Allocation in distributed Clouds

Location-Aware Multi-user Resource Allocation in Distributed...

引用

10th Annual Conference of Advanced Computer Architecture, ACA 2014

作者： Li, Jiaxin Li, Dongsheng Zheng, Jing Quan, Yong National Key Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha China Information Center of Logistics Department Beijing China School of Computer Science National University of Defense Technology Changsha China

ISBN: (纸本)9783662444900

Resource allocation for multi-user across multiple data centers is an important problem in cloud computing environments. Many geographically-distributed users may request virtualized resources simultaneously. And the distances from users to allocated resources have much impact on the quality of service (QoS) in multiple data centers environment. Most existing methods do not take all these factors into account when allocating resources. They usually result in poor runtime performance of users' virtual computing environment and the remarkable difference of users' QoS. In this paper, we propose RAMD, a resource allocation algorithm based on multi-stage decision in multiple data centers. The RAMD algorithm allocate VMs to users, taking into account the correlation and interaction between multiple users, so as to minimize the sum of all users' service distances (i.e. determined by user location and network distance of virtual machines). Experimental results show that the algorithm can effectively deal with the cloud resource allocation for multi-user across multiple data centers. It can improve the runtime performance of users' virtualized resources and reduce the difference of QoS. © Springer-Verlag Berlin Heidelberg 2014.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

MLPs: Efficient Training of MiniGo on Large-scale Heterogeneous Computing System

MLPs: Efficient Training of MiniGo on Large-scale Heterogene...

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Peng Qiao Zhouyu He Rongchun Li Jingfei Jiang Yong Dou Dongsheng Li Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

Deep Reinforcement Learning has been successfully applied in various applications and achieved impressive performance compared with previous traditional methods but suffers from high computation cost and long training time. MLPerf takes deep reinforcement learning as one of the benchmark tracks and provides a single node training version of MiniGo as a reference. A key challenge is to achieve efficient MiniGo training on a large-scale computing system. According to the training computation pattern in MiniGo and the characteristics of our large-scale heterogeneous computing system, we propose a MultiLevel parallel strategy, MLPs, including task-level parallelism between nodes, CPU-DSP heterogeneous parallelism, and DSP multi-core parallelism. The proposed method reduces the overall execution time from 43 hours to 16 hours while scaling the node size from 1067 to 4139. The scaling efficiency is 69.1%. According to our fitting method, the scaling efficiency is 46.5% when scaling to 8235 nodes. The experimental results show that the proposed method achieves the efficient training of MiniGo on the largescale heterogeneous computing system.

关键词： Training Deep learning Neural networks Memory management Reinforcement learning parallel processing Programming

来源：评论

学校读者我要写书评

暂无评论

Aircraft Segmentation Based On Deep Learning framework : from extreme points to remote sensing image segmentation

Aircraft Segmentation Based On Deep Learning framework : fro...

引用

IEEE Symposium Series on Computational Intelligence (SSCI)

作者： Lei Zhao Peng Qiao Yong Dou Science and Technology on Parallel and Distributed Laboratory National University of Defense Changsha China

ISBN: (数字)9781728124858

ISBN: (纸本)9781728124865

Remote sensing image segmentation is a very important technology. Although the segmentation method based on convolutional neural networks (CNNs) has achieved promising results in natural image test set, e.g. VOC or COCO, they provide inferior performance when being transferred to remote sensing images. Due to the limits of labeled remote sensing images, finetuning pre-trained CNNs using remote sensing images do not benefit the image segmentation performance. Inspired by the recent works of interactive segmentation methods which exploit several extreme clicks that are fed into CNNs to improve the accuracy of the segmentation, we propose an effective method to improve the segmentation accuracy, which uses four extreme points (the top, bottom, left, and right) as the guide information. In terms of mIoU, our method achieves 84.4% on remote sensing image dataset, which outperforms the previous work by 23.1%. Compared with the previous interactive segmentation methods, the proposed method achieves superior performance. In addition, an improved method with an extra point is proposed based on the inaccurate part of results obtained by four extreme points. It is very feasible to be applied in an interactive segmentation toolbox.

关键词： Image segmentation Remote sensing Training Aircraft Perturbation methods Semantics Heating systems

来源：评论

学校读者我要写书评

暂无评论

Diagonalwise refactorization: An efficient training method for depthwise convolutions

arXiv

引用

arXiv 2018年

作者： Qin, Zheng Zhang, Zhaoning Li, Dongsheng Zhang, Yiming Peng, Yuxing Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

Depthwise convolutions provide significant performance benefits owing to the reduction in both parameters and mult-adds. However, training depthwise convolution layers with GPUs is slow in current deep learning frameworks because their implementations cannot fully utilize the GPU capacity. To address this problem, in this paper we present an efficient method (called diagonalwise refactorization) for accelerating the training of depthwise convolution layers. Our key idea is to rearrange the weight vectors of a depthwise convolution into a large diagonal weight matrix so as to convert the depthwise convolution into one single standard convolution, which is well supported by the cuDNN library that is highly-optimized for GPU computations. We have implemented our training method in five popular deep learning frameworks. Evaluation results show that our proposed method gains 15.4× training speedup on Darknet, 8.4× on Caffe, 5.4× on PyTorch, 3.5× on MXNet, and 1.4× on TensorFlow, compared to their original implementations of depthwise convolutions. Copyright © 2018, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Merging and evolution: Improving convolutional neural networks for mobile applications

arXiv

引用

arXiv 2018年

作者： Qin, Zheng Zhang, Zhaoning Zhang, Shiqing Yu, Hao Peng, Yuxing Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

Compact neural networks are inclined to exploit "sparsely-connected" convolutions such as depthwise convolution and group convolution for employment in mobile applications. Compared with standard "fully-connected" convolutions, these convolutions are more computationally economical. However, "sparsely-connected" convolutions block the inter-group information exchange, which induces severe performance degradation. To address this issue, we present two novel operations named merging and evolution to leverage the inter-group information. Our key idea is encoding the inter-group information with a narrow feature map, then combining the generated features with the original network for better representation. Taking advantage of the proposed operations, we then introduce the Merging-and-Evolution (ME) module, an architectural unit specifically designed for compact networks. Finally, we propose a family of compact neural networks called MENet based on ME modules. Extensive experiments on ILSVRC 2012 dataset and PASCAL VOC 2007 dataset demonstrate that MENet consistently outperforms other state-of-the-art compact networks under different computational budgets. For instance, under the computational budget of 140 MFLOPs, MENet surpasses ShuffleNet by 1% and MobileNet by 1.95% on ILSVRC 2012 top-1 accuracy, while by 2.3% and 4.1% on PASCAL VOC 2007 mAP, respectively. Copyright © 2018, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

FD-Mobilenet: Improved mobilenet with a fast downsampling strategy

arXiv

引用

arXiv 2018年

作者： Qin, Zheng Zhang, Zhaoning Chen, Xiaotao Peng, Yuxing Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology Changsha China

We present Fast-Downsampling MobileNet (FD-MobileNet), an efficient and accurate network for very limited computational budgets (e.g., 10-140 MFLOPs). Our key idea is applying a fast downsampling strategy to MobileNet framework. In FD-MobileNet, we perform 32× downsampling within 12 layers, only half the layers in the original MobileNet. This design brings three advantages: (i) It remarkably reduces the computational cost. (ii) It increases the information capacity and achieves significant performance improvements. (iii) It is engineering-friendly and provides fast actual inference speed. Experiments on ILSVRC 2012 and PASCAL VOC 2007 datasets demonstrate that FD-MobileNet consistently outperforms MobileNet and achieves comparable results with ShuffleNet under different computational budgets, for instance, surpassing MobileNet by 5.5% on the ILSVRC 2012 top-1 accuracy and 3.6% on the VOC 2007 mAP under a complexity of 12 MFLOPs. On an ARM-based device, FD-MobileNet achieves 1.11× inference speedup over MobileNet and 1.82× over ShuffleNet under the same complexity. Copyright © 2018, The Authors. All rights reserved.

关键词： Signal sampling

来源：评论

学校读者我要写书评

暂无评论

Topology-aware deployment of scientific applications in cloud computing

Topology-aware deployment of scientific applications in clou...

引用

2012 IEEE 5th International Conference on Cloud Computing, CLOUD 2012

作者： Fan, Pei Chen, Zhenbang Wang, Ji Zheng, Zibin Lyu, Michael R. National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China Dept. of Computer Science and Engineering Chinese University of Hong Kong Hong Kong Hong Kong

ISBN: (纸本)9780769547558

Nowadays, more and more scientific applications are moving to cloud computing. The optimal deployment of scientific applications is critical for providing good services to users. Scientific applications are usually topology-aware applications. Therefore, considering the topology of a scientific application during the development will benefit the performance of the application. However, it is challenging to automatically discover and make use of the communication pattern of a scientific application while deploying the application on cloud. To attack this challenge, in this paper, we propose a framework to discover the communication topology of a scientific application by pre-execution and multi-scale graph clustering, based on which the deployment can be optimized. Comprehensive experiments are conducted by employing a well-known MPI benchmark and comparing the performance of our method with those of other methods. The experimental results show the effectiveness of our topology-aware deployment method. © 2012 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Toward optimal deployment of communication-intensive cloud applications

Toward optimal deployment of communication-intensive cloud a...

引用

International Conference on Cloud Computing

作者： Fan, Pei Wang, Ji Zheng, Zibin Lyu, Michael R. National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China Dept. of Computer Science and Engineering Chinese University of Hong Kong Hong Kong Hong Kong

ISBN: (纸本)9780769544601

Strongly promoted by the leading industrial companies, cloud computing becomes increasingly popular in recent years. The growth rate of cloud computing surpasses even the most optimistic predictions. A cloud application is a large-scale distributed system that consist a lot of distributed cloud nodes. How to make optimal deployment of cloud applications is a challenging research problem. When deploying a cloud application to the cloud environment, cloud node ranking is one of the most important approaches for selecting optimal cloud nodes for the cloud application. Traditional ranking methods usually rank the cloud nodes based on their QoS values, without considering the communication performance between cloud nodes. However, such kind of node relationship is very important for the communication-intensive cloud applications (e.g., Message Passing Interface (MPI) programs), which have a lot of communications between the selected cloud nodes. In this paper, we propose a novel clustering-based method for selecting optimal cloud nodes for deploying communication-intensive applications to the cloud environment. Our method not only takes into account the cloud node qualities, but also the communication performance between different nodes. We deploy several well-known MPI programs on a real-world cloud and compare our method with other methods. The experimental results show the effectiveness of our cluster-based method. © 2011 IEEE.

关键词： Quality of service

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：