检索结果-内蒙古大学图书馆

European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD)

作者： Nolet, Corey J. Gala, Divye Fender, Alex Doijade, Mahesh Eaton, Joe Raff, Edward Zedlewski, John Rees, Brad Oates, Tim NVIDIA Inc Santa Clara CA 95051 USA Univ Maryland Baltimore Cty Baltimore MD 21228 USA Booz Allen Hamilton Mclean VA USA

ISBN: (纸本)9783031434112;9783031434129

In this paper, we propose cuSLINK, a novel and state-of-the-art reformulation of the SLINK algorithm on the GPU which requires only O(Nk) space and uses a parameter k to trade off space and time. We also propose a set of novel and reusable building blocks that compose cuSLINK. These building blocks include highly optimized computational patterns for k-NN graph construction, spanning trees, and dendrogram cluster extraction. We show how we used our primitives to implement cuSLINK end-to-end on the GPU, further enabling a wide range of real-world data mining and machine learning applications that were once intractable. In addition to being a primary computational bottleneck in the popular HDBSCAN algorithm, the impact of our end-to-end cuSLINK algorithm spans a large range of important applications, including cluster analysis in social and computer networks, natural language processing, and computer vision.

关键词： KNN Graph Neighborhood Methods Nearest Neighbors Spanning Tree Single-Linkage Hierarchical Clustering Agglomerative Clustering Cluster Analysis Networks Forest parallel algorithms GPU

来源：评论

学校读者我要写书评

暂无评论

A new parallel adaptive clustering and its application to streaming data

引用

JOURNAL OF COMPUTATIONAL SCIENCE 2023年 66卷

作者： McLaughlin, Benjamin Ha Kang, Sung Asbury Univ Shaw Sch Sci 1 Macklem Dr Wilmore KY 40390 USA Georgia Inst Technol Sch Math 686 Cherry St NW Atlanta GA 30332 USA

This paper presents a parallel adaptive clustering (PAC) algorithm to automatically classify data while simultaneously choosing a suitable number of classes. Clustering is an important tool for data analysis and understanding in a broad set of areas including data reduction, pattern analysis, and classification. However, the requirement to specify the number of clusters in advance and the computational burden associated with clustering large sets of data persist as challenges in clustering. We propose a new parallel adaptive clustering (PAC) algorithm that addresses these challenges by adaptively computing the number of clusters and leveraging the power of parallel computing. The algorithm clusters disjoint subsets of the data on parallel computation threads. We develop regularized set k-means to efficiently cluster the results from the parallel threads. A refinement step further improves the clusters. The PAC algorithm offers the capability to adaptively cluster data sets which change over time by reusing the information from previous time steps to decrease computation. We provide theoretical analysis and numerical experiments to characterize the performance of the method, validate its properties, and demonstrate the computational efficiency of the method.

关键词： Unsupervised clustering k-means parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

A multi-level parallel algorithm for seismic imaging based on one-way wave equation migration

引用

COMPUTERS & MATHEMATICS WITH APPLICATIONS 2021年 97卷 344-354页

作者： Pleshkevich, Alexander Lisitsa, Vadim Vishnevsky, Dmitry Levchenko, Vadim Cent Geophys Expedit JSC Moscow Russia RAS SB Inst Math Novosibirsk Russia RAS Inst Petr Geol & Geophys SB Novosibirsk Russia Keldysh Inst Appl Math Moscow Russia

This article presents a parallel algorithm for seismic imaging based on depth wavefield extrapolation (the solution of a one-way wave equation [OWE]) employing the pseudospectral method. The algorithm is essentially oriented on common-offset vector (COV) gathers seismic migration based on an OWE, includes several parallelism levels, and utilizes message passing interface (MPI), Nested OpenMP, and CUDA technologies. The uppermost level of parallelism involves data splitting to ensure that each dataset can be processed independently to construct parts of the COV images. The algorithm is embarrassingly parallel at this level. Next, each independent run processes several datasets to compute COV images. Each MPI process computes all COV images for a single dataset, then MPI processes exchange data to build up a single COV image for all datasets at a single node. Computation of the COV image at a node requires OpenMP parallelization so that one thread governs the GPU-based calculations and facilitates the construction of images within a thick slab. All other threads perform image interpolation within the slab. Finally, CUDA technology is used for the most computationally intense part of the algorithm-wavefield extrapolation. Such a complex algorithm structure makes it possible to process all COV images for full-azimuth seismic data employing OWE-based migration.

关键词： One-way wave equation Pseudo-spectral method Seismic imaging GPU parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

PiP-MColl: Process-in-Process-based Multi-object MPI Collectives 25

PiP-MColl: Process-in-Process-based Multi-object MPI Collect...

引用

25th IEEE International Conference on Cluster Computing (CLUSTER)

作者： Huang, Jiajun Ouyang, Kaiming Zhai, Yujia Liu, Jinyang Si, Min Raffenetti, Ken Zhou, Hui Hori, Atsushi Chen, Zizhong Guo, Yanfei Thakur, Rajeev Univ Calif Riverside Riverside CA 92521 USA NVIDIA Corp Santa Clara CA USA Meta Platforms Inc Cambridge MA USA Argonne Natl Lab Argonne IL USA Natl Inst Informat Chiyoda City Tokyo Japan

ISBN: (纸本)9798350307924

In the era of exascale computing, the adoption of a large number of CPU cores and nodes by high-performance computing (HPC) applications has made MPI collective performance increasingly crucial. As the number of cores and nodes increases, the importance of optimizing MPI collective performance becomes more evident. Current collective algorithms, including kernel-assisted inter-process data exchange techniques and data sharing based shared-memory approaches, are prone to significant performance degradation due to the overhead of system calls and page faults or the cost of extra data-copy latency. These issues can negatively impact the efficiency and scalability of HPC applications. To address these issues, we propose PiP-MColl, a Process-in-Process-based Multi-object Interprocess MPI Collective design that maximizes small message MPI collective performance at scale. We also present specific designs to boost the performance for larger messages, such that we observe a comprehensive improvement for a series of message sizes beyond small messages. PiP-MColl features efficient multiple sender and receiver collective algorithms and leverages Processin-Process shared memory techniques to eliminate unnecessary system call, page fault overhead and extra data copy, which results in improved intra- and inter-node message rate and throughput. Experimental results demonstrate that PiP-MColl significantly outperforms popular MPI libraries, including OpenMPI, MVAPICH2, and Intel MPI, by up to 4.6X for the MPI collectives MPI Scatter, MPI Allgather, and MPI Allreduce.

关键词： MPI Collective Message Passing Interface Process-in-Process parallel algorithms Distributed Systems

来源：评论

学校读者我要写书评

暂无评论

Property Graphs in Arachne

Property Graphs in Arachne

引用

IEEE High Performance Extreme Computing Virtual Conference (HPEC)

作者： Rodriguez, Oliver Alvarado Buschmann, Fernando Vera Du, Zhihui Bader, David A. New Jersey Inst Technol Dept Data Sci Newark NJ 07102 USA

ISBN: (纸本)9798350308600

Analyzing large-scale graphs poses challenges due to their increasing size and the demand for interactive and user-friendly analytics tools. These graphs arise from various domains, including cybersecurity, social sciences, health sciences, and network sciences, where networks can represent interactions between humans, neurons in the brain, or malicious flows in a network. Exploring these large graphs is crucial for revealing hidden structures and metrics that are not easily computable without parallel computing. Currently, Python users can leverage the open-source Arkouda framework to efficiently execute Pandas and NumPy-related tasks on thousands of cores. To address large-scale graph analysis, Arachne, an extension to Arkouda, enables easy transformation of Arkouda dataframes into graphs. This paper proposes and evaluates three distributable data structures for property graphs, implemented in Chapel, that are integrated into Arachne. Enriching Arachne with support for property graphs will empower data scientists to extend their analysis to new problem domains. Property graphs present additional complexities, requiring efficient storage for extra information on vertices and edges, such as labels, relationships, and properties.

关键词： graph analytics parallel algorithms property graphs distributed-memory

来源：评论

学校读者我要写书评

暂无评论

Fantastic Subgraphs and How to Find Them

Fantastic Subgraphs and How to Find Them

引用

作者： Sukprasert, Pattara Northwestern University

学位级别：Ph.D., Doctor of Philosophy

Due to their widespread applicability, graphs and networks appear in various contexts. The increasing scale of graphs encountered in the real-world requires the development of efficient algorithms that run reasonably fast and produce close to optimal solutions. The main focus of this thesis is the development of fast graph algorithms for optimizing specific structural properties. Many classical problems can be thought of in this framework - for example, the maximum weight matching problem in graphs can be thought of as finding a maximum weight subgraph where each node has degree at most one. While these problems are known to be solvable in polynomial time, many interesting properties could lead to NP-hard optimization problems. We study four problems that fit in the framework. First, we study k-Edge Connected Spanning Subgraph (kECSS). Our goal here is to come up with a network that tolerates link failures. Given an undirected weighted graph, we want to find a minimum weighted spanning subgraph with a property that this subgraph must remain connected even after removing k edges. This problem has applications in route planning where links such as roads or network cables might become unusable. We can think of it as a generalization to the classical minimum spanning tree problem, and this belongs to a family of problems called Survivable Network Design. It has been known for a long time how to get a 2-approximation solution in polynomial time. We show how to find a fractional solution in near-linear time and how to round it to an integral solution efficiently. Second, we study the Dynamic Graph Spanner problem. Here, our goal is to develop a data structure, called a spanner, that maintains a subgraph approximating distances between pairs of nodes dynamically. A dynamic graph undergoes structural changes such as edge insertions and deletions. More precisely, our goal is to maintain a t-spanner, which is small subgraph such that the distance between every pair of vertice

关键词： algorithms Approximation algorithms Combinatorial optimization Dynamic algorithms Graph theory parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

An adaptive parallel evolutionary algorithm for solving the uncapacitated facility location problem

引用

EXPERT SYSTEMS WITH APPLICATIONS 2023年 224卷

作者： Sonuc, Emrullah Ozcan, Ender Univ Nottingham Sch Comp Sci Computat Optimisat & Learning COL Lab Nottingham NG8 1BB England Karabuk Univ Fac Engn Dept Comp Engn TR-78050 Karabuk Turkiye

Metaheuristics, providing high level guidelines for heuristic optimisation, have successfully been applied to many complex problems over the past decades. However, their performances often vary depending on the choice of the initial settings for their parameters and operators along with the characteristics of the given problem instance handled. Hence, there is a growing interest into designing adaptive search methods that automate the selection of efficient operators and setting of their parameters during the search process. In this study, an adaptive binary parallel evolutionary algorithm, referred to as ABPEA, is introduced for solving the uncapacitated facility location problem which is proven to be an NP-hard optimisation problem. The approach uses a unary and two other binary operators. A reinforcement learning mechanism is used for assigning credits to operators considering their recent impact on generating improved solutions to the problem instance in hand. An operator is selected adaptively with a greedy policy for perturbing a solution. The performance of the proposed approach is evaluated on a set of well-known benchmark instances using ORLib and M*, and its scaling capacity by running it with different starting points on an increasing number of threads. Parameters are adjusted to derive the best configuration of three different rewarding schemes, which are instant, average and extreme. A performance comparison to the other state-of-the-art algorithms illustrates the superiority of ABPEA. Moreover, ABPEA provides up to a factor of 3.9 times acceleration when compared to the sequential algorithm based on a single-operator.

关键词： Adaptive operator selection Metaheuristics Combinatorial optimisation parallel algorithms

来源：评论

学校读者我要写书评

暂无评论

HermesBDD: A Multi-Core and Multi-Platform Binary Decision Diagram Package 26

HermesBDD: A Multi-Core and Multi-Platform Binary Decision D...

引用

26th International Symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)

作者： Capogrosso, Luigi Geretti, Luca Cristani, Marco Fummi, Franco Villa, Tiziano Univ Verona Dept Comp Sci Verona Italy

ISBN: (纸本)9798350332773

BDDs are representations of a Boolean expression in the form of a directed acyclic graph. BDDs are widely used in several fields, particularly in model checking and hardware verification. There are several implementations for BDD manipulation, where each package differs depending on the application. This paper presents HermesBDD: a novel multi-core and multi-platform binary decision diagram package focused on high performance and usability. HermesBDD supports a static and dynamic memory management mechanism, the possibility to exploit lock-free hash tables, and a simple parallel implementation of the IF-THEN - ELSE procedure based on a higher-level wrapper for threads and futures. HermesBDD is completely written in C++ with no need to rely on external libraries and is developed according to software engineering principles for reliability and easy maintenance over time. We provide experimental results on the n-Queens problem, the de-facto SAT solver benchmark for BDDs, demonstrating a significant speedup of 18.73x over our non-parallel baselines, and a remarkable performance boost w.r.t. other state-of-the-art BDDs packages.

关键词： Binary Decision Diagrams Boolean Functions parallel algorithms Multi-Platform Package

来源：评论

学校读者我要写书评

暂无评论

Improving Efficiency of parallel Vertex-Centric algorithms for Irregular Graphs

引用

IEEE TRANSACTIONS ON parallel AND DISTRIBUTED SYSTEMS 2019年第10期30卷 2265-2282页

作者： Ozdal, Muhammet Mustafa Bilkent Univ Dept Comp Engn TR-06800 Ankara Turkey

Memory access is known to be the main bottleneck for shared-memory parallel graph applications especially for large and irregular graphs. Propagation blocking (PB) idea was proposed recently to improve the parallel performance of PageRank and sparse matrix and vector multiplication operations. The idea is based on separating parallel computation into two phases, binning and accumulation, such that random memory accesses are replaced with contiguous accesses. In this paper, we propose an algorithm that allows execution of these two phases concurrently. We propose several improvements to increase parallel throughput, reduce memory overhead, and improve work efficiency. Our experimental results show that our proposed algorithms improve shared-memory parallel throughput by a factor of up to 2x compared to the original PB algorithms. We also show that the memory overhead can be reduced significantly (from 170 percent down to less than 5 percent) without significant degradation of performance. Finally, we demonstrate that our concurrent execution model allows asynchronous parallel execution, leading to significant work efficiency in addition to throughput improvements.

关键词： parallel algorithms graph algorithms sparse matrix vector multiplication (SpMV) sparse matrix sparse vector multiplication (SpMSpV) high performance computing

来源：评论

学校读者我要写书评

暂无评论

parallel Median Filter with Arbitrary Window Size and Image Depth

Parallel Median Filter with Arbitrary Window Size and Image ...

引用

Euromicro Conference on parallel, Distributed and Network-Based Processing

作者： Moreno Marzolla Michele Ravaioli Elena Loli Piccolomini Department of Computer Science and Engineering Center for Inter-Department Industrial Research ICT University of Bologna Bologna Italy Department of Computer Science and Engineering University of Bologna Bologna Italy

ISBN: (数字)9798331524937

ISBN: (纸本)9798331524944

Removing noise in digital images is a fundamental operation that arises in many application domains. In this paper we consider the median filter, a filtering technique that replaces the color of each pixel with the median of those in a square neighborhood of fixed radius. For some use cases, the size of the neighborhood or the image depth may be large, making existing algorithms either too slow, or not applicable at all due to excessive memory requirements. In this paper we describe architecture-specific optimizations that enable the computation of the median filter with arbitrary window size and image depth on multicore processors and GPUs. We report preliminary results that indicate that the parallel implementations are suitable for practical use, with the GPU version outperforming the CPU.

关键词： Histograms Single instruction multiple data Multicore processing Image color analysis Memory management Graphics processing units Filtering algorithms Programming parallel algorithms Optimization

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：