检索结果-内蒙古大学图书馆

computing in Science and Engineering 2001年第6期3.0卷 97-99页

作者： Zomaya, Albert Y. Parallel Computing Research Laboratory Dept. of Electrical and Electronic Eng. University of Western Australia Nedlands WA Australia

Not Available

关键词： Simulated annealing Cost function Temperature Random number generation Optimization methods Data structures Space exploration Processor scheduling Scheduling algorithm Mirrors

来源：评论

学校读者我要写书评

暂无评论

Powerlyra: Differentiated graph computation and partitioning on skewed graphs 15

Powerlyra: Differentiated graph computation and partitioning...

引用

10th European Conference on Computer Systems, EuroSys 2015

作者： Chen, Rong Shi, Jiaxin Chen, Yanzhe Chen, Haibo Shanghai Key Laboratory of Scalable Computing and Systems Institute of Parallel and Distributed Systems Shanghai Jiao Tong University China

ISBN: (纸本)9781450332385

Natural graphs with skewed distribution raise unique challenges to graph computation and partitioning. Existing graph-parallel systems usually use a "one size fits all" design that uniformly processes all vertices, which either suffer from notable load imbalance and high contention for high-degree vertices (e.g., Pregel and GraphLab), or incur high communication cost and memory consumption even for low-degree vertices (e.g., PowerGraph and GraphX). In this paper, we argue that skewed distribution in natural graphs also calls for differentiated processing on highdegree and low-degree vertices. We then introduce Power-Lyra, a new graph computation engine that embraces the best of both worlds of existing graph-parallel systems, by dynamically applying different computation and partitioning strategies for different vertices. PowerLyra further provides an efficient hybrid graph partitioning algorithm (hybrid-cut) that combines edge-cut and vertex-cut with heuristics. Based on PowerLyra, we design locality-conscious data layout optimization to improve cache locality of graph accesses during communication. PowerLyra is implemented as a separate computation engine of PowerGraph, and can seamlessly support various graph algorithms. A detailed evaluation on two clusters using graph-analytics and MLDM (machine learning and data mining) applications show that PowerLyra outperforms PowerGraph by up to 5.53X (from 1.24X) and 3.26X (from 1.49X) for real-world and synthetic graphs accordingly, and is much faster than other systems like GraphX and Giraph, yet with much less memory consumption. A porting of hybrid-cut to GraphX further confirms the efficiency and generality of PowerLyra. Copyright © 2015 ACM 978-1-4503-3238-5/15/04 $15.00.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

parallel discrete element simulation of a heterogeneous particle system

Parallel discrete element simulation of a heterogeneous part...

引用

5th International Conference on Engineering Computational Technology, ECT 2006

作者： Kačianauskas, R. Maknickas, A. Kačeniauskas, A. Markauskas, D. Balevičius, R. Parallel Computing Laboratory Vilnius Gediminas Technical University Vilnius Lithuania Laboratory of Numerical Modelling Vilnius Gediminas Technical University Vilnius Lithuania

ISBN: (纸本)1905088094

The paper describes parallel 3D DEM simulation of compacting of heterogeneous material. Static domain decomposition and message passing inter-processor communication have been implemented in the DEM code. A novel algorithm for moving particles, that exchange processors, is incorporated in the domain decomposition framework. The speed-up equal to 8.81 has been obtained on 10 processors of the distributed memory PC cluster. It has been observed that computational efficiency is sensitive to increase of heterogeneity. The developed parallel algorithm and software are applied to compacting of granular material by rigid walls. Simulation results are presented in terms of wall pressures, the coordination number and the packing density. © 2006 Civil-Comp Press.

关键词： parallel processing systems

来源：评论

学校读者我要写书评

暂无评论

Using JIT compilation and configurable runtime systems for efficient deployment of java programs on ubiquitous devices

引用

International Conference on Ubiquitous computing, Ubicomp 2001

作者： Teodorescu, Radu Pandey, Raju Parallel and Distributed Computing Laboratory Computer Science Department University of California Davis United States

ISBN: (纸本)3540426140

As the proliferation of ubiquitous devices moves computation away from the conventional desktop computer boundaries, distributed systems design is being exposed to new challenges. A distributed system supporting a ubiquitous computing application must deal with a wider spectrum of hardware architectures featuring structural and functional differences, and resources limitations. Due to its architecture independent infrastructure and object-oriented programming model, the Java programming environment can support flexible solutions for addressing the diversity among these devices. Unfortunately, Java solutions are often associated with high costs in terms of resource consumption, which limits the range of devices that can benefit from this approach. In this paper, we present an architecture that deals with the cost and complexity of running Java programs by partitioning the process of Java program execution between system nodes and remote devices. The system nodes prepare a Java application for execution on a remote device by generating device-specific native code and application-specific runtime system on the fly. The resulting infrastructure provides the flexibility of a high-level programming model and the architecture independence specific to Java. At the same time the amount of resources consumed by an application on the targeted device are comparable to that of a native implementation. © Springer-Verlag Berlin Heidelberg 2001.

关键词： Object oriented programming

来源：评论

学校读者我要写书评

暂无评论

A comparative study of star graphs and rotator graphs

A comparative study of star graphs and rotator graphs

引用

23rd International Conference on parallel Processing, ICPP 1994

作者： Ponnuswamy, Subburajan Chaudhary, Vipin Parallel and Distributed Computing Laboratory Dept. of ECE Wayne State University Detroit MI 48202 United States

ISBN: (纸本)0849324939

Star graph is an extensively studied Cayley graph, considered to be an attractive alternative to the popular binary cube. The rotator graphs are a set of directed Cayley graphs introduced recently. In this paper we compare the structural and algorithmic aspects of star graphs with that of rotator graphs. In the process we present some new results for star graphs and rotator graphs. We present a formula for the number of nodes at any distance from the identity permutation in star graphs. The minimum bisection width of star and rotator graphs is obtained. Partitioning and fault tolerant parameters for both star and rotator graphs are analyzed. The node disjoint parallel paths and hence the upper bound on the fault diameter of rotator graphs are presented. We compare the minimal path routing in star and rotator graphs using simulation results. © 1994 IEEE.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Mbapp: Efficient Memory-Balanced Pipeline parallelism for Large Model Fine-Tuning on Commodity GPU Servers 24

Mbapp: Efficient Memory-Balanced Pipeline Parallelism for La...

引用

5th International Conference on Computer Information and Big Data Applications, CIBDA 2024

作者： Liu, Yujie Lai, Zhiquan Li, Dongsheng National Key Laboratory of Parallel and Distributed Computing College of Computer National University of Defense Technology Changsha410000 China

ISBN: (纸本)9798400718106

Large-scale models have demonstrated outstanding performance across various downstream tasks. Pipeline parallelism is essential for fine-tuning large models on commodity GPU servers, as it plays a crucial role in making their exceptional performance more accessible and widespread. Existing approaches have encountered challenges in attaining effective memory-balanced pipeline parallelism. This paper presents a novel memory load-balanced pipeline parallel solution that aims to distribute memory usage evenly across stages on commodity GPU servers by leveraging NVLink bridges. The solution presents a novel approach for transferring data from GPUs to CPUs through the PCI-e link between adjacent GPUs interconnected by the NVLink bridge. Moreover, our methodology orchestrates data transfer operations to reduce offloading latency during the fine-tuning of large models. The evaluation demonstrates that our approach enhances switching efficiency by 1.46x to 1.76x and improves throughput by 2.04x to 3.59x compared to the PyTorch offloading technique. © 2024 ACM.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Literature recommendation based on reference graph

Literature recommendation based on reference graph

引用

International Conference on Advanced Computer Theory and Engineering

作者： Yan, Yang Long, Yun Key Laboratory of Database and Parallel Computing of Heilongjiang Province Department of Computer Science and Technology Heilongjiang University Harbin China

ISBN: (纸本)9781424465408

Research papers are frequently requested in digital library. When users query for documents about a research problem, users usually want all the related documents and the citations with the relations among them, to enable the user know the development of the research area. Traditional recommendation techniques are mainly based on similarity among users or documents to do recommendation. They do not use the relationships among documents and, especially, cannot provide topology structure of related documents to the user. This paper proposes a new recommendation technique based on reference graph of documents. Combined with traditional recommendation techniques, it can recommend research paper more completely, accurately and clearly. Experimental results show that the reference graph based recommendation technique is more effective. © 2010 IEEE.

关键词： Digital libraries

来源：评论

学校读者我要写书评

暂无评论

Influence Maximization in Microblog Social Network 7

Influence Maximization in Microblog Social Network

引用

7th International Conference on Information Science and Control Engineering, ICISCE 2020

作者： Ma, Zhengyi Yang, Yan Heilongjiang University School of Computer Science Technology Harbin China Key Laboratory of Database and Parallel Computing Harbin China

ISBN: (纸本)9781728164069

In recent years, the continuous rise of social networks has promoted the research of maximizing influence. Therefore, it is more and more important to mine a group of users with the largest influence range in social network, which is also of great significance for the promotion of some products. To solve the above problems, this paper studies the influence maximization in microblog social network, and proves that the problem is NP hard, and the objective function of the problem satisfies monotonicity and submodule. An effective algorithm is proposed to solve this problem. The algorithm fully considers the nature of microblog social network and microblog users, and measures the influence of users themselves with activity, communication power and coverage. Based on the independent cascade model, a communication model is designed according to the influence of users and the interaction times between users. Through the experiment on the real microblog data set, it is proved that the algorithm has good performance in the propagation range and operation efficiency, and the propagation effect is more in line with the reality. © 2020 IEEE.

关键词： Economic and social effects

来源：评论

学校读者我要写书评

暂无评论

Identifying faults in large-scale distributed systems by filtering noisy error logs

Identifying faults in large-scale distributed systems by fil...

引用

Workshop on Proactive Failure Avoidance, Recovery, and Maintenance

作者： Rao, Xiang Wang, Huaimin Shi, Dianxi Chen, Zhenbang Cai, Hua Zhou, Qi Sun, Tingtao National Laboratory for Parallel and Distributed Processing National University of Defense Technology Changsha 410073 China Computing Platform Alibaba Cloud Computing Corporation Hangzhou China

ISBN: (纸本)9781457703751

Extracting fault features with the error logs of fault injection tests has been widely studied in the area of large scale distributed systems for decades. However, the process of extracting features is severely affected by a large amount of noisy logs. While the existing work tries to solve the problem by compressing logs in temporal and spatial views or removing the semantic redundancy between logs, they fail to consider the co-existence of other noisy faults that generate error logs instead of injected faults, for example, random hardware faults, unexpected bugs of softwares, system configuration faults or the error rank of a log severity. During a fault feature extraction process, those noisy faults generate error logs that are not related to a target fault, and will strongly mislead the resulted fault features. We call an error log that is not related to a target fault a noisy error log. To filter out noisy error logs, we present a similarity-based error log filtering method SBF, which consists of three integrated steps: (1) model error logs into time series and use haar wavelet transform to get the approximate time series;(2) divide the approximate time series into sub time series by valleys;(3) identify noisy error logs by comparing the similarity between the sub time series of target error logs and the template of noisy error logs. We apply our log filtering method in an enterprise cloud system and show its effectiveness. Compared with the existing work, we successfully filter out noisy error logs and increase the precision and the recall rate of fault feature extraction.1 © 2011 IEEE.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Sensitivity loss training based implicit feedback 27

Sensitivity loss training based implicit feedback

引用

27th IEEE International Conference on parallel and Distributed Systems, ICPADS 2021

作者： Li, Kun Wang, Nan Liu, Xinyu School of Computer Science and Technology Heilongjiang University Key Laboratory of Database and Parallel Computing of Heilongjiang Province Harbin China

ISBN: (纸本)9781665408783

In recommender systems, due to the lack of explicit feedback features, datasets with implicit feedback are always accustomed to train all samples without separating them during model training, without considering the non-consistency of samples. This leads to a significant decrease in sample utilization and creates challenges for model training. Also, little work has been done to explore the intrinsic laws implied in the implicit feedback dataset and how to effectively train the implicit feedback data. In this paper, we first summarize the variation pattern of loss with model training for different rating samples in the explicit feedback dataset, and find that model training is highly sensitive to the ratings. Second, we design an adaptive hierarchical training function with dynamic thresholds that can effectively distinguish different rating samples in the dataset, thus optimizing the implicit feedback dataset into an explicit feedback dataset to some extent. Finally, to better learn samples with different ratings, we also propose an adaptive hierarchical training strategy to obtain better training results in the implicit feedback dataset. Extensive experiments on three datasets show that our approach achieves excellent performance and greatly improves the performance of the model. © 2021 IEEE.

关键词： adaptive hierarchical training implicit feedback sample utilization sensitive

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：