检索结果-内蒙古大学图书馆

International Conference on High Performance Computing in the Asia-Pacific Region

作者： Yonggang Che Zhenghua Wang Xiaomei Li National Laboratory for Parallel and Distributed Processing Changsha China

Program performance optimization often involves choosing right parameters to minimize the program's runtime. Selecting optimization parameters by means of execution-driven search is guaranteed to find excellent results, for it accurately accounts for all performance components of the target platform. But the major drawback of execution-driven approach is the excessive compilation time due to thousands of runs of the original program. In this article, we propose a novel technique called program reduction transformations to reduce the cost of execution-driven optimization parameter selection. It is based on our observation to the characteristics of the scientific applications and the optimization parameter selection task. The ideal is to transform the program before it is used in execution-driven parameter selection procedure. The transformed program runs in much shorter time but preserves the parameter selection quality. This technique greatly reduces the time spent on evaluating each candidate parameter and makes execution-driven optimization parameter selection affordable. We formulate the theoretic foundation of program reduction transformation. And we find several situations where reduction transformations can be legally applied. These situations are common in scientific applications. Experiments done for two math kernels and three SPEC benchmarks show that our approach is both feasible and effective

关键词： Tiles Runtime Cost function Predictive models Hardware distributed processing Kernel Optimization methods parallel processing Data analysis

来源：评论

学校读者我要写书评

暂无评论

Verification of C programs using slicing execution

Verification of C programs using slicing execution

引用

International Conference on Quality Software

作者： Xiaodong Yi Ji Wang Xuejun Yang National Laboratory for Parallel and Distributed Processing Changsha China

The paper presents a novel method, namely slicing execution, to verify C programs with respect to temporal safety properties. The distinguished feature is that it only simulates the execution of the relevant statements under abstraction criteria and checks the properties on the fly. The abstraction criterion begins with a proper initial set of program variables and may be iteratively refined according to spurious counter-examples. Provided that the properties to be verified usually involve only a few variables in practical programs, slicing execution may have the same precision as path-sensitive simulation with the cost close to standard flow-sensitive dataflow analysis. The presented method has been used to verify the initial handshake process of SSL protocol based on the C source code of openssl-0.9.6c. The experiment results confirm our claim and show that slicing execution is not only practical but also effective.

关键词： Safety Scalability Costs Data analysis Analytical models Protocols Laboratories distributed processing Guidelines Performance analysis

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

2010 International Conference in Information Science and Applications, ICISA 2010

作者： Fu, Hongyi Ding, Yan Key Laboratory of Science and Technology for National Defence of Parallel and Distributed Processing National University of Defence Tech. Changsha Hunan 410073 China School of Computer National University of Defence Tech. Changsha Hunan 410073 China

ISBN: (纸本)9781424459438

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. ©2010 IEEE.

关键词： Fault tolerance

来源：评论

学校读者我要写书评

暂无评论

Revised stride data value predictor design

Revised stride data value predictor design

引用

International Conference on High Performance Computing in the Asia-Pacific Region

作者： Xiao Yong Yang Yanping Zhou Xingming National Laboratory for Parallel and Distributed Processing Changsha China

Stride data value predictor is widely used by researchers in data value prediction study. Compared with context-based hybrid data value predictors, stride data value predictors are simple. But when encountering non-stride repeated sequences, a stride value predictor does not perform as well as a context-based hybrid data value predictor. In this paper, a revised stride data value predictor is introduced. With a little augment to a traditional stride data value predictor, the new predictor can make correct predictions on some patterns that can only be done by the context-based data value predictors. Simulation results show that the new predictor works well with most value predictable instructions. Design decisions such as predictor size, confidence mechanism and storing partial tag are analyzed

关键词： Predictive models History Counting circuits Laboratories distributed processing Costs Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Contrastive Learning with Diverse Samples 26

Contrastive Learning with Diverse Samples

引用

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Wu, Lilei Liu, Jie Science and Technology on Parallel and Distributed Processing Laboratory National University of Defense Technology Changsha410073 China Laboratory of Digitizing Software for Frontier Equipment National University of Defense Technology Changsha410073 China

ISBN: (纸本)9781643684369

Unsupervised visual representation learning has gained much attention from the computer vision community because of the recent contrastive learning achievements. Current work mainly adopts instance discrimination as the pretext task, which treats every single instance as a different class (negative) and uses a collection of data augmentation techniques to generate more examples (positive) for each class. The idea is straightforward and efficient but will generally cause similar instances to be classified into different classes. Such problem has been defined as 'class collision' in some previous works and is shown to hurt the representation ability. Motivated by this observation, we present a solution to address this issue by filtering similar negative examples from each mini-batch. Concretely, we model the problem as a Determinantal Point Process (DPP) so that similar instances can be filtered stochastically, and diverse samples are expected to be sampled for contrastive training. Besides, we further introduce a priority term for each instance, which indicates the hardness of its positives, so that instances with more hard positives are more likely to be sampled for contributing to the optimization. Our sampling can be efficiently implemented in a feed-forward manner and further accelerated by our encouraged complement DPP. Extensive experimental results demonstrate our priority over the standard setup of contrastive learning. © 2023 The Authors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Overview of blockchain consensus mechanism 2020

Overview of blockchain consensus mechanism

引用

2nd International Conference on Big Data Engineering, BDE 2020

作者： Zhang, Changqiang Wu, Cangshuai Wang, Xinyi Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology China College of Computer National University of Defense Technology China

ISBN: (纸本)9781450377225

The advent of the Bitcoin system has brought another boom in the Internet era. In a very short time, many Blockchain systems come into being successively, whose decentration, consensus mechanisms, intelligent contract, and other characteristics make them applicable to various fields such as finance, education, medical, technology, etc. The consensus mechanism is the core of Blockchain technology. And a good consensus mechanism plays a very important role in the stable operation of the Blockchain system. The continuous improvement of consensus mechanisms such as PoW (proof-of-work), PoS (proof-of-stake), DPoS (delegated-proof-of-stake), and PBFT (Practical Byzantine Fault Tolerance) has led to the evolution of Blockchain technology to Blockchain 3.0. Starting from the issue of Byzantine generals, this article analyzes common consensus mechanisms based on existing Blockchain applications and then evaluates their consistency. This paper has a unique perspective, selection and comment on the latest progress and important literature in the field of block chain. At last, it also has a high degree of summary and prospect to the existing problems and the future development trend. © 2020 ACM.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Mobility of Internet-Based Virtual Computing Environment

Mobility of Internet-Based Virtual Computing Environment

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Siqi Shen Ji Wang Rui Shen Shengdong Zhang Pei Fan National Laboratory for Parallel and Distributed Processing Changsha China

The Internet-based virtual computing environment (iVCE) provides on-demand aggregation and autonomic collaboration mechanisms to facilitate the utilization of autonomous and dynamic Internet resources. Load balancing and fault tolerance are important issues when scheduling those transient resources. In this paper, we propose a mobility mechanism for the migration of various roles of agents in the iVCE platform. The mobility mechanism involves two parts of the iVCE platform: role container layer and event service layer. At the role container layer, a novel approach is proposed to handle the code and data mobility issue. At the event service layer, an efficient routing reconfiguration protocol is proposed based on a publish/subscribe system over DHTs to facilitate task migrations. Certain conditions must be satisfied before the migration of an agent to ensure the correctness of the whole process. Experiments are conducted to evaluate the performance of the mobility mechanism, and the experimental results show that it is suitable for implementing load balancing and fault tolerance in the iVCE.

关键词： Internet Load management Fault tolerance Containers distributed computing Computer languages Concurrent computing International collaboration Routing protocols Middleware

来源：评论

学校读者我要写书评

暂无评论

A survey on approaches and applications of knowledge representation learning 5

A survey on approaches and applications of knowledge represe...

引用

5th IEEE International Conference on Data Science in Cyberspace, DSC 2020

作者： Li, Chenchen Li, Aiping Wang, Ye Tu, Hongkui Song, Yichen National University of Defense Technology College of Computer ChangSha China National University of Defense Technology College of Computer Science and Technology on Parallel and Distributed Processing Laboratory ChangSha China

ISBN: (纸本)9781728195582

Knowledge representation learning (KRL) is one of the important research topics in artificial intelligence and Natural language processing. It can efficiently calculate the semantics of entities and relations in a low-dimensional space, and effectively solve the problem of data sparsity, which significantly improve the performance of knowledge acquisition, fusion and reasoning and so on. Starting from the five perspectives of distance-based, semantic matching, bilinear-based, neural network model and additional information model, this paper first introduces the overall framework and specific model design, and then correspondingly introduces the experimental evaluation tasks, metrics and benchmark datasets of each model. On this basis, how to apply KRL to various downstream tasks is summarized. © 2020 IEEE.

关键词： Knowledge representation

来源：评论

学校读者我要写书评

暂无评论

Towards a Framework for Scalable Model Checking of Concurrent C Programs

Towards a Framework for Scalable Model Checking of Concurren...

引用

International Symposium on Leveraging Applications of Formal Methods, Verification and Validation ( ISoLA)

作者： Ji Wang Xiaodong Yi Xuejun Yang National Laboratory for Parallel and Distributed Processing Changsha China

The paper presents a novel framework for scalable model checking of concurrent C programs. With the idea of verification reuse, it shows an integrated approach to efficient reduction of state space by abstraction, symbolic representation and dynamic partial-order reduction (DPOR) techniques. The framework is founded on an over-approximated model of the concurrent program by variable abstraction, and combines DPOR with lightweight symbolic execution to generate the symbolic conditions for all locations, called -conditions, which are intended for verification reuse. The -conditions of a location are weak approximation of the conditions that must be satisfied at that location so as to guarantee the temporal safety properties to be verified. These conditions will be checked for reusing the previous exploration in verification, and will be iteratively refined under the guidance of spurious counterexamples. The presented framework is demonstrated by several experiments including a concurrent software system whose server and client processes are derived from openssl-0.9.6c C source codes implementing the SSL protocol.

关键词： State-space methods Safety Open source software Java Laboratories distributed processing Space exploration Interleaved codes Scalability Software systems

来源：评论

学校读者我要写书评

暂无评论

Cooperative monitoring BGP among autonomous systems

引用

Security and Communication Networks 2015年第10期8卷 1943-1957页

作者： Hu, Ning Wang, BaoSheng Liu, Xin National Key Laboratory for Parallel and Distributed Processing College of Computer National University of Defense Technology Changsha Hunan China

As the de facto Internet inter-domain routing protocol, BGP protocol has a number of vulnerabilities and weakness. Monitoring BGP is an effective way to improve the security of inter-domain routing. This paper presents a cooperative BGP monitoring method, which is called cooperative information sharing model (CoISM). CoISM is based on self-organization and can be used to solve some problems during BGP monitoring, such as route validation and bogus route notification delivery. CoISM provides autonomous systems with a more comprehensive information view by introducing information diffuse reflection based on initiative inquiry and making use of the relativity of monitoring information. CoISM optimizes the information transmission by leveraging the data locality caused by BGP policy and implements ISP coordination with low communication and deployment cost. More specifically, CoISM provides a self-organizing and incentive mechanism, which drives autonomous systems to coordinate independently and shares information on-demand. CoISM supports incremental deployment and can also be applied to a wide range of inter-domain cooperative management applications such as inter-domain routing failure analysis and intrusion detection. © 2014 John Wiley & Sons, Ltd.

关键词： Monitoring

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：