检索结果-内蒙古大学图书馆

Data management for large‐scale scientific computations in high performance distributed systems

Cluster computing 2000年第1期3卷 45-60页

作者： Choudhary, A. Kandemir, M. No, J. Memik, G. Shen, X. Liao, W. Nagesh, H. More, S. Taylor, V. Thakur, R. Stevens, R. Center for Parallel and Distributed Computing Department of Electrical and Computer Engineering Northwestern University Evanston USA Department of Computer Science and Engineering The Pennsylvania State University University Park USA Mathematics and Computer Science Division Argonne National Laboratory Argonne USA

With the increasing number of scientific applications manipulating huge amounts of data, effective high-level data management is an increasingly important problem. Unfortunately, so far the solutions to the high‐level data management problem either require deep understanding of specific storage architectures and file layouts (as in high-performance file storage systems) or produce unsatisfactory I/O performance in exchange for ease-of-use and portability (as in relational DBMSs). In this paper we present a novel application development environment which is built around an active meta-data management system (MDMS) to handle high-level data in an effective manner. The key components of our three-tiered architecture are user application, the MDMS, and a hierarchical storage system (HSS). Our environment overcomes the performance problems of pure database-oriented solutions, while maintaining their advantages in terms of ease-of-use and portability. The high levels of performance are achieved by the MDMS, with the aid of user-specified, performance-oriented directives. Our environment supports a simple, easy-to-use yet powerful user interface, leaving the task of choosing appropriate I/O techniques for the application at hand to the MDMS. We discuss the importance of an active MDMS and show how the three components of our environment, namely the application, the MDMS, and the HSS, fit together. We also report performance numbers from our ongoing implementation and illustrate that significant improvements are made possible without undue programming effort.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Memory service based memory sharing grid system

引用

Jisuanji Xuebao/Chinese Journal of Computers 2006年第7期29卷 1225-1233页

作者： Chu, Rui Xiao, Nong Lu, Xi-Cheng National Key Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China

Many scientific computing applications demand a great amount of memory, which are usually run on supercomputers with large physical memory. The applications whose complete time is affected by memory capacity, is usually called memory-intensive applications. Many disk 10 operations will be caused by memory-intensive applications due to insufficient physical memory of computing environment, which depress the system performance dramatically. Traditional network memory or memory server schemes try to share free memory of idle nodes in the cluster, while it is often influenced by heavy cluster loading or interior network congestion. Combining the network memory with current service computing and grid computing technology, a memory service based memory sharing grid system named RAM Grid, is proposed. RAM grid improve traditional network memory scheme, and extend the application area of grid computing. As one of the major computing resources, memory can be shared on the Internet through RAM grid, such as computing power and disk storage being shared through computational grid and data grid. The system criteria and properties are analyzed. One of the system schemes is also proposed in detail. Through the real trace based simulation, the performance of RAM Grid is evaluated and proved better than traditional network memory or disk IO obviously.

关键词： Random access storage

来源：评论

学校读者我要写书评

暂无评论

Deep Isolation Forest for Anomaly Detection

arXiv

引用

arXiv 2022年

作者： Xu, Hongzuo Pang, Guansong Wang, Yijie Wang, Yongjun The College of Computer National University of Defense Technology Changsha410073 China The Science and Technology on Parallel and Distributed Processing Laboratory China The School of Computing and Information Systems Singapore Management University 178902 Singapore

Isolation forest (iForest) has been emerging as arguably the most popular anomaly detector in recent years due to its general effectiveness across different benchmarks and strong scalability. Nevertheless, its linear axis-parallel isolation method often leads to (i) failure in detecting hard anomalies that are difficult to isolate in high-dimensional/non-linear-separable data space, and (ii) notorious algorithmic bias that assigns unexpectedly lower anomaly scores to artefact regions. These issues contribute to high false negative errors. Several iForest extensions are introduced, but they essentially still employ shallow, linear data partition, restricting their power in isolating true anomalies. Therefore, this paper proposes deep isolation forest. We introduce a new representation scheme that utilises casually initialised neural networks to map original data into random representation ensembles, where random axis-parallel cuts are subsequently applied to perform the data partition. This representation scheme facilitates high freedom of the partition in the original data space (equivalent to non-linear partition on subspaces of varying sizes), encouraging a unique synergy between random representations and random partition-based isolation. Extensive experiments show that our model achieves significant improvement over state-of-the-art isolation-based methods and deep detectors on tabular, graph and time series datasets;our model also inherits desired scalability from iForest. © 2022, CC BY.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

SATMask: Spatial Attention Transform Mask for Dense Instance Segmentation 7

SATMask: Spatial Attention Transform Mask for Dense Instance...

引用

7th IEEE International Conference on Data Science in Cyberspace, DSC 2022

作者： Mao, Quanzhong Sun, Lijuan Wu, Jingchen Gao, Yutong Wu, Xu Qiu, Lirong Ministry of Education Key Laboratory of Trustworthy Distributed Computing and Service China Bupt School of Cyberspace Security Bupt School of Computer Science National Pilot Software Engineering School Beijing Jiaotong University Beijing University of Posts and Telecommunications Beijing China Beijing University of Posts and Telecommunications Library China

ISBN: (数字)9781665474801

ISBN: (纸本)9781665474801

There are often dense objects in the images processed by instance segmentation, but too dense objects will cause the problem that the objects are difficult to segment. Most of the current dense instance segmentation methods are based on dense sliding window such as TensorMask. However, the sliding window has the problems of high computation and difficult design of anchor. In order to solve the above difficulties, we propose an anchor-free and single shot dense image segmentation framework, named SATMask, which adds a Spatial Attention Transform (SAT) mask head on anchor-free one stage object detector (FCOS) to predict high quality instance mask with low complexity, and uses feature-aligned pyramid network to fuse the feature map generated by backbone to obtain rich spatial details and better semantic information. Extensive experiments on the challenging COCO and Cityscapes datasets demonstrate the effectiveness of SATMask. In particular, under the same backbone (ResNet-101), SATMask achieves 39.8% AP on COCO, surpassing the state-of-the-art instance segmentation method Mask R-CNN 1.5% AP. © 2022 IEEE.

关键词： Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

Cooperative Air-Ground Instant Delivery by UAVs and Crowdsourced Taxis

Cooperative Air-Ground Instant Delivery by UAVs and Crowdsou...

引用

International Conference on Data Engineering

作者： Junhui Gao Qianru Wang Xin Zhang Juan Shi Xiang Zhao Qingye Han Yan Pan School of Computer Science Northwestern Polytechnical University School of Computer Science and Technology Xidian University National Key Laboratory of Information Systems Engineering National University of Defense Technology Air Force Engineering University Laboratory for Big Data and Decision National University of Defense Technology School of Management Science and Real Estate Chongqing University National Key Laboratory of Parallel and Distributed Computing National University of Defense Technology

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Instant delivery has become a fundamental service in people's daily lives. Different from the traditional express service, the instant delivery has a strict shipping time constraint after being ordered. However, the labor shortage makes it challenging to realize efficient instant delivery. To tackle the problem, researchers have studied to introduce vehicles (i.e., taxis) or Unmanned Aerial Vehicles (UAVs or drones) into instant delivery tasks. Unfortunately, the delivery detour of taxis and the limited battery of UAVs make it hard to meet the rapidly increasing instant delivery demands. Under this circumstance, this paper proposes an air-ground cooperative instant delivery paradigm to maximize the delivery performance and meanwhile minimize the negative effects on the taxi passengers. Specifically, a data-driven delivery potential-demands-aware cooperative strategy is designed to improve the overall delivery performance of both UAVs and taxis as well as the taxi passengers' experience. The experimental results show that the proposed method improves the delivery number by 30.1% and 114.5% compared to the taxi-based and UAV-based instant delivery respectively, and shortens the delivery time by 35.7% compared to the taxi-based instant delivery.

关键词： Autonomous aerial vehicles Data engineering Air to ground communication Trajectory Batteries Time factors Task analysis

来源：评论

学校读者我要写书评

暂无评论

distributed Relation Discovery in Internet of Things

Distributed Relation Discovery in Internet of Things

引用

International Conference on Cloud computing and Big Data (CCBD)

作者： Weiping Zhu Hongliang Lu Xiaohui Cui International School of Software Wuhan University China Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology China Department of Computing Hong Kong Polytechnic University Hong Kong

ISBN: (纸本)9781479966226

Internet of things is an emerging technology that aims to connect various smart objects in our daily life. It facilitates the information exchange and control among objects. In the Internet of things, it is important to discover various relations among objects for analyzing and mining useful knowledge. Existing works on relation discovery mainly focus on centralized processing. It is not suitable for Internet of things due to unavailable of server, one-point failure, computation bottleneck, and security and business concerns. In this paper, we propose a distributed approach to discover the relations among objects in Internet of things. We first build the distritbuted system model which may include multiple relation discovery tasks. Based on that, we design an approach utilizing distributed spanning tree to extract the relations. Rather than sending all the information to the server, the objects only need to send the information to a local leader object in our algorithm. We also discuss more about how to improve the performance of the proposed approach and how to relax the system constraints. Extensive simulation have been done and the results show that the proposed approach outperforms existing approaches in terms of the data amount of transmission.

关键词： Internet of things Synchronization Buildings Servers Algorithm design and analysis Delays

来源：评论

学校读者我要写书评

暂无评论

A Multi-Objective Emergency Resource Scheduling Method Based on MOEA/D 19

A Multi-Objective Emergency Resource Scheduling Method Based...

引用

Proceedings of the 2019 2nd International Conference on Algorithms, computing and Artificial Intelligence

作者： Yanrong Li Xiaoyong Li Liyang Hou Wenping Kong Wenxue Ma Key Laboratory of Trustworthy Distributed Computing and Servic Beijing University of Posts and Telecommunications Beijing China National Engineering Laboratory For Integrated Command And Dispatch Technology Beijing China

ISBN: (纸本)9781450372619

After unconventional events occur, rapid emergency response is often needed in order to reduce losses. When there are more than one disaster points and more than one supply points, how to coordinate the local dispatch of emergency resources among these points is a problem worth studying. In this paper, we design a multi-objective, multi-point to multi-point and multi-resource emergency scheduling model that consider resources arrive time and transportation cost. Then a new emergency resource scheduling algorithm based on multi objective evolutionary algorithm based on decomposition (MOEA/D) is proposed, a new mutation operator was also designed to solve this problem. The experimental results show that this method can get a feasible scheduling scheme, which is faster than NSGA-II and has better scheduling results.

关键词： MOEA/D

来源：评论

学校读者我要写书评

暂无评论

Using Redundant Threads for Fault Tolerance of OpenMP Programs

Using Redundant Threads for Fault Tolerance of OpenMP Progra...

引用

International Conference on Information Science and Applications (ICISA)

作者： Hongyi Fu Yan Ding Key Laboratory of Science and Technology for National Defence of Parallel and Distributed Processing National University of Defense Technology Changsha Hunan China Section #620 School of Computer National University of Defense Technology Changsha Hunan China

ISBN: (纸本)9781424459421

As the wide application of multi-core processor architecture in the domain of high performance computing, fault tolerance for shared memory parallel programs becomes a hot spot of research. For years, checkpointing has been the dominant fault tolerance technology in this field, and recently, many research works have been engaged with it. However, to those programs which deal with large amount of data, checkpointing may induce massive I/O transfer, which will adversely affect scalability. To deal with such a problem, this paper proposes a fault tolerance approach, making use of redundancy, for shared memory parallel programs. Our scheme avoids saving and restoring computational state during the program's execution, hence does not involve I/O operations, so presents explicit advantage over checkpointing in scalability. In this paper, we introduce our approach and the related compiler tool in detail, and give the experimental evaluation result.

关键词： Fault tolerance Checkpointing Scalability Hardware Application software Multicore processing High performance computing Laboratories distributed processing

来源：评论

学校读者我要写书评

暂无评论

HPDedup: A hybrid prioritized data deduplication mechanism for primary storage in the cloud

arXiv

引用

arXiv 2017年

作者： Wu, Huijun Wang, Chen Fu, Yinjin Sakr, Sherif Zhu, Liming Lu, Kai Data61 CSIRO University of New South Wales Australia PLA University of Science and Technology China Science and Technology on Parallel and Distributed Laboratory State Key Laboratory of High Performance Computing State Key Laboratory of High-end Server & Storage Technology College of Computer National University of Defense Technology Changsha China

—Eliminating duplicate data in primary storage of clouds increases the cost-efficiency of cloud service providers as well as reduces the cost of users for using cloud services. Most existing primary deduplication techniques either use inline caching to exploit locality in primary workloads or use post-processing deduplication running in system idle time to avoid the negative impact on I/O performance. However, neither of them works well in the cloud servers running multiple services or applications for the following two reasons: Firstly, the temporal locality of duplicate data writes may not exist in some primary storage workloads thus inline caching often fails to achieve good deduplication ratio. Secondly, the post-processing deduplication allows duplicate data to be written into disks, therefore does not provide the benefit of I/O deduplication and requires high peak storage capacity. This paper presents HPDedup, a Hybrid Prioritized data Deduplication mechanism to deal with the storage system shared by applications running in co-located virtual machines or containers by fusing an inline and a post-processing process for exact deduplication. In the inline deduplication phase, HPDedup gives a fingerprint caching mechanism that estimates the temporal locality of duplicates in data streams from different VMs or applications and prioritizes the cache allocation for these streams based on the estimation. HPDedup also allows different deduplication threshold for streams based on their spatial locality to reduce the disk fragmentation. The post-processing phase removes duplicates whose fingerprints are not able to be cached due to weak temporal locality from disks. The hybrid deduplication mechanism significantly reduces the amount of redundant data written to the storage system while maintaining inline data writing performance. Our experimental results show that HPDedup clearly outperforms the state-of-the-art primary storage deduplication techniques in terms of inline

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Vision-Language Navigation: A Survey and Taxonomy

arXiv

引用

arXiv 2021年

作者： Wu, Wansen Chang, Tao Li, Xinmeng Yin, Quanjun Zhu, Jiancheng The College of Systems Engineering National University of Defense Technology Hunan Changsha410073 China National Key Laboratory of Parallel and Distributed Processing College of Computer Science and Technology National University of Defence Technology Changsha410073 China

Vision-Language Navigation (VLN) tasks require an agent to follow human language instructions to navigate in previously unseen environments. This challenging field involving problems in natural language processing, computer vision, robotics, etc., has spawn many excellent works focusing on various VLN tasks. This paper provides a comprehensive survey and an insightful taxonomy of these tasks based on the different characteristics of language instructions in these tasks. Depending on whether the navigation instructions are given for once or multiple times, this paper divides the tasks into two categories, i.e., single-turn and multi-turn tasks. For single-turn tasks, we further subdivide them into goal-oriented and route-oriented based on whether the instructions designate a single goal location or specify a sequence of multiple locations. For multi-turn tasks, we subdivide them into passive and interactive tasks based on whether the agent is allowed to question the instruction or not. These tasks require different capabilities of the agent and entail various model designs. We identify progress made on the tasks and look into the limitations of existing VLN models and task settings. Finally, we discuss several open issues of VLN and point out some opportunities in the future, i.e., incorporating knowledge with VLN models and implementing them in the real physical world. © 2021, CC BY.

关键词： Taxonomies

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：