检索结果-内蒙古大学图书馆

FMCC-RT: a scalable and fine-grained all-reduce algorithm for large-scale SMP clusters

Science China(Information Sciences) 2025年第5期68卷 362-379页

作者： Jintao PENG Jie LIU Jianbin FANG Min XIE Yi DAI Zhiquan LAI Bo YANG Chunye GONG Xinjun MAO Guo MAO Jie REN School of Computer Science and Technology National University of Defense Technology Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Laboratory of Digitizing Software for Frontier Equipment National University of Defense Technology National Supercomputer Center in Tianjin School of Computer Science Shaanxi Normal University

All-reduce is a widely used communication technique for distributed and parallel applications typically implemented using either a tree-based or ring-based scheme. Each of these approaches has its own limitations: tree-based schemes struggle with efficiently exchanging large messages, while ring-based solutions assume constant communication throughput,an unrealistic expectation in modern network communication infrastructures. We present FMCC-RT, an all-reduce approach that combines the advantages of tree-and ring-based implementations while mitigating their drawbacks. FMCC-RT dynamically switches between tree and ring-based implementations depending on the size of the message being processed. It utilizes an analytical model to assess the impact of message sizes on the achieved throughput, enabling the derivation of optimal work partitioning parameters. Furthermore, FMCC-RT is designed with an Open MPI-compatible API, requiring no modification to user code. We evaluated FMCC-RT through micro-benchmarks and real-world application tests. Experimental results show that FMCC-RT outperforms state-of-the-art tree-and ring-based methods, achieving speedups of up to 5.6×.

关键词： all-reduce collective communication MPI scalability

来源：评论

学校读者我要写书评

暂无评论

YFLM: An Improved Levenberg-Marquardt Algorithm for Global Bundle Adjustment 41st

YFLM: An Improved Levenberg-Marquardt Algorithm for Global ...

引用

41st computer Graphics International Conference, CGI 2024

作者： Peng, Jiaxin Li, Tao Jiang, Qin Liu, Jie Wang, Ruibo Laboratory of Software Engineering for Complex Systems School of Computer Science National University of Defense Technology Hunan Changsha410073 China Parallel and Distributed Processing Laboratory School of Computer Science National University of Defense Technology Hunan Changsha410073 China

ISBN: (纸本)9783031820205

The conventional Levenberg-Marquardt (LM) algorithm is a state-of-the-art trust-region optimization method for solving bundle adjustment problems in the Structure-from-Motion community, which not only takes advantage of the fast convergence of the Gauss-Newton method, but also the stability of the gradient descent method when approaching optimal solutions. However, the damping ratio of LM is simply provided by trial-and-error, which causes slow convergence rate for large-scale problems. This paper proposes the Yamashita-Fukushima LM (YFLM) algorithm to reduce the time complexity for global bundle adjustment, where the damping factor is determined by Yamashita and Fukushima’s method. YFLM dynamically calculates a more reasonable and optimal damping ratio according to the newest reprojection error. The experimental results show that the YFLM algorithm outperforms the conventional LM algorithm for most public bundle adjustment datasets. Besides this, the convergence of the YFLM algorithm is also evaluated with different σ∈(0,2]. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Damping

来源：评论

学校读者我要写书评

暂无评论

AFMA-Track: Adaptive Fusion of Motion and Appearance for Robust Multi-object Tracking 27th

AFMA-Track: Adaptive Fusion of Motion and Appearance for ...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Liao, Wei Luo, Lei Zhang, Chunyuan College of Computer Science and Technology National University of Defence Technology Changsha China Science and Technology on Parallel and Distributed Processing Laboratory College of Computer Science and Technology National University of Defense Technology Changsha China

ISBN: (纸本)9783031784439

Motion and appearance cues play a crucial role in Multi-object Tracking (MOT) algorithms for associating objects across consecutive frames. While most MOT methods prioritize accurate motion modeling and distinctive appearance representations, the use of appearance and motion cues is often confined to simplistic association techniques. For instance, fixed weights are commonly employed to combine the intersection-over-union (IoU) matrix and appearance similarity matrix, yielding an association cost matrix. To harness the full potential of motion and appearance cues across diverse scenarios, we propose an innovative approach that dynamically balances motion and appearance cues based on scene and object information during the association process. Furthermore, we introduce a new mechanism for updating appearance representations, effectively mitigating noise introduced by occlusion. Our method demonstrates state-of-the-art performance on the MOT17 and MOT20 test sets. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Deadlock Prevention for GPU Collective Communication 25

Comprehensive Deadlock Prevention for GPU Collective Communi...

引用

20th European Conference on computer Systems, EuroSys 2025, co-located 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, ASPLOS 2025

作者： Pan, Lichen Liu, Juncheng Fu, Yongquan Yuan, Jinhui Zhang, Rongkai Li, Pengze Xiao, Zhen School of Computer Science Peking University China OneFlow Research China National Key Laboratory of Parallel and Distributed Computing College of Computer Science and Technology National University of Defense Technology China

ISBN: (纸本)9798400711961

distributed deep neural network training necessitates efficient GPU collective communications, which are inherently susceptible to deadlocks. GPU collective deadlocks arise easily in distributed deep learning applications when multiple collectives circularly wait for each other. GPU collective deadlocks pose a significant challenge to the correct functioning and efficiency of distributed deep learning, and no general effective solutions are currently available. Only in specific scenarios, ad-hoc methods, making an application invoke collectives in a consistent order across GPUs, can be used to prevent circular collective dependency and deadlocks. This paper presents DFCCL, a novel GPU collective communication library that provides a comprehensive approach for GPU collective deadlock prevention while maintaining high performance. DFCCL achieves preemption for GPU collectives at the bottom library level, effectively preventing deadlocks even if applications cause circular collective dependency. DFCCL ensures high performance with its execution and scheduling methods for collectives. Experiments show that DFCCL effectively prevents GPU collective deadlocks in various situations. Moreover, extensive evaluations demonstrate that DFCCL delivers performance comparable to or superior to NCCL, the state-of-the-art collective communication library highly optimized for NVIDIA GPUs. © 2025 Copyright held by the owner/author(s).

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

VirtMan:design and implementation of a fast booting system for homogeneous virtual machines in iVCE

引用

Frontiers of Information Technology & Electronic Engineering 2016年第2期17卷 110-121页

作者： Zi-yang LI Yi-ming ZHANG Dong-sheng LI Peng-fei ZHANG Xi-cheng LU National Laboratory for Parallel and Distributed Processing School of ComputerNational University of Defense Technology

Internet-based virtual computing environment （iVCE） has been proposed to combine data centers and other kinds of computing resources on the Internet to provide efficient and economical services. Virtual machines （VMs） have been widely used in iVCE to isolate different users/jobs and ensure trustworthiness, but traditionally VMs require a long period of time for booting, which cannot meet the requirement of iVCE＇s large-scale and highly dynamic applications. To address this problem, in this paper we design and implement VirtMan, a fast booting system for a large number of virtual machines in iVCE. VirtMan uses the Linux Small computer System Interface （SCSI） target to remotely mount to the source image in a scalable hierarchy, and leverages the homogeneity of a set of VMs to transfer only necessary image data at runtime. We have implemented VirtMan both as a standalone system and for OpenStack. In our 100-server testbed, VirtMan boots up 1000 VMs （with a 15 CB image of Windows Server 2008） on 100 physical servers in less than 120 s, which is three orders of magnitude lower than current public clouds.

关键词： Virtual machine Fast booting Homogeneity Internet-based virtual computing environment （iVCE）

来源：评论

学校读者我要写书评

暂无评论

Managing Data-Objects in Dynamically Reconfigurable Caches

引用

Journal of computer Science & Technology 2010年第2期25卷 232-245页

作者：杨学军吴俊杰曾坤唐玉华 National Laboratory for Parallel and Distributed Processing School of ComputerNational University of Defense Technology

The widening gap between processor and memory speeds makes cache an important issue in the computer system design. Compared with work set of programs, cache resource is often rare. Therefore, it is very important for a computer system to use cache efficiently. Toward a dynamically reconfigurable cache proposed recently, DOOC （Data- Object Oriented Cache）, this paper proposes a quantitative framework for analyzing the cache requirement of data-objects, which includes cache capacity, block size, associativity and coherence protocol. And a kind of graph coloring algorithm dealing with the competition between data-objects in the DOOC is proposed as well. Finally, we apply our approaches to the compiler management of DOOC. We test our approaches on both a single-core platform and a four-core platform. Compared with the traditional caches, the DOOC in both platforms achieves an average reduction of 44.98% and 49.69% in miss rate respectively. And its performance is very close to the ideal optimal cache.

关键词： cache data-object data reuse data-object oriented cache （DOOC） graph coloring

来源：评论

学校读者我要写书评

暂无评论

Determinants of pull-based development in the context of continuous integration

引用

Science China(Information Sciences) 2016年第8期59卷 53-66页

作者： Yue YU Gang YIN Tao WANG Cheng YANG Huaimin WANG College of Computer National University of Defense Technology National Laboratory for Parallel and Distributed Processing

The pull-based development model, widely used in distributed software teams on open source communities, can efficiently gather the wisdom from crowds. Instead of sharing access to a central repository,contributors create a fork, update it locally, and request to have their changes merged back, i.e., submit a pull-request. On the one hand, this model lowers the barrier to entry for potential contributors since anyone can submit pull-requests to any repository, but on the other hand it also increases the burden on integrators, who are responsible for assessing the proposed patches and integrating the suitable changes into the central repository. The role of integrators in pull-based development is crucial. They must not only ensure that pull-requests should meet the project’s quality standards before being accepted, but also finish the evaluations in a timely manner. To keep up with the volume of incoming pull-requests, continuous integration(CI) is widely adopted to automatically build and test every pull-request at the time of submission. CI provides extra evidences relating to the quality of pull-requests, which would help integrators to make final decision(i.e., accept or reject). In this paper, we present a quantitative study that tries to discover which factors affect the process of pull-based development model, including acceptance and latency in the context of CI. Using regression modeling on data extracted from a sample of Git Hub projects deploying the Travis-CI service, we find that the evaluation process is a complex issue, requiring many independent variables to explain adequately. In particular, CI is a dominant factor for the process, which not only has a great influence on the evaluation process per se, but also changes the effects of some traditional predictors.

关键词： pull-request continuous integration Git Hub distributed software development empirical analysis

来源：评论

学校读者我要写书评

暂无评论

MABP: an optimal resource allocation approach in data center networks

引用

Science China(Information Sciences) 2014年第10期57卷 230-245页

作者： LI XiaoLing WANG HuaiMin DING Bo LI XiaoYong National Key Laboratory for Parallel and Distributed Processing School of ComputerNational University of Defense Technology

In data center networks, resource allocation based on workload is an effective way to allocate the infrastructure resources to diverse cloud applications and satisfy the quality of service for the users, which refers to mapping a large number of workloads provided by cloud users/tenants to substrate network provided by cloud providers. Although the existing heuristic approaches are able to find a feasible solution, the quality of the solution is not guaranteed. Concerning this issue, based on the minimum mapping cost, this paper solves the resource allocation problem by modeling it as a distributed constraint optimization problem. Then an efficient approach is proposed to solve the resource allocation problem, aiming to find a feasible solution and ensuring the optimality of the solution. Finally, theoretical analysis and extensive experiments have demonstrated the effectiveness and efficiency of our proposed approach.

关键词： data center network resource allocation workload substrate network optimality distributed constraint optimization

来源：评论

学校读者我要写书评

暂无评论

The influences of model parameters on the characteristics of memristors

引用

Chinese Physics B 2012年第4期21卷 576-585页

作者：周静黄达 National Laboratory for Parallel and Distributed Processing School of ComputerNational University of Defense TechnologyChangsha 410073China

As the fourth passive circuit component, a memristor is a nonlinear resistor that can ＂remember＂ the amount of charge passing through it. The characteristic of ＂remembering＂ the charge and non-volatility makes memristors great potential candidates in many fields. Nowadays, only a few groups have the ability to fabricate memristors, and most researchers study them by theoretic analysis and simulation. In this paper, we first analyse the theoretical base and characteristics of memristors, then use a simulation program with integrated circuit emphasis as our tool to simulate the theoretical model of memristors and change the parameters in the model to see the influence of each parameter on the characteristics. Our work supplies researchers engaged in memristor-based circuits with advice on how to choose the proper parameters.

关键词： memristor I-V characteristics simulation program with integrated circuit emphasis

来源：评论

学校读者我要写书评

暂无评论

Feverfew: a scalable coverage-based hybrid overlay for Internet-scale pub/sub networks

引用

Science China(Information Sciences) 2014年第5期57卷 51-64页

作者： MA XingKong WANG YiJie SUN WeiDong National Key Laboratory for Parallel and Distributed Processing School of ComputerNational University of Defense Technology

The publish/subscribe(pub/sub)paradigm is a popular communication model for data dissemination in large-scale distributed ***,scalability comes with a contradiction between the delivery latency and the memory *** one hand,constructing a separate overly per topic guarantees real-time dissemination,while the number of node degrees rapidly increases with the number of *** the other hand,maintaining a bounded number of connections per node guarantees small memory cost,while each message has to traverse a large number of uninterested nodes before reaching the *** this paper,we propose Feverfew,a coverage-based hybrid overlay that disseminates messages to all subscribers without uninterested nodes involved in,and increases the average number of node connections slowly with an increase in the number of subscribers and *** major novelty of Feverfew lies in its heuristic coverage mechanism implemented by combining a gossip-based sampling protocol with a probabilistic searching *** on the practical workload,our experimental results show that Feverfew significantly outperforms existing coverage-based overlay and DHT-based overlay in various dynamic network environments.

关键词： publish/subscribe coverage-based data dissemination topic-based gossip

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：