检索结果-内蒙古大学图书馆

A large-scale heterogeneous computing framework for non-uniform sampling two-dimensional convolution applications

CCF TRANSACTIONS ON HIGH PERFORMANCE COMPUTING 2024年第2期6卷 221-239页

作者： Lu, Yu Yu, Ce Xiao, Jian Wang, Hao Fu, Hao Kang, Bo Zheng, Gang Tianjin Univ Coll Intelligence & Comp Tianjin Peoples R China Natl Supercomp Ctr Tianjin Tianjin Peoples R China

Non-uniform sampling two-dimensional convolution (NUSC) maps spatially sampling data with irregular distribution to a regular grid by convolution. As the data scale and growth rate continue to increase, accelerating NUSC with the heterogeneous computing platform is a feasible way. However, the complex hardware architecture and storage hierarchy of the heterogeneous computing platform poses a challenge to programming and performance tuning. Therefore, this paper proposes a heterogeneous parallel programming model and runtime framework named AutoNUSC. For the programming difficulties of NUSC in heterogeneous computing environments, AutoNUSC abstracts and encapsulates the parallel execution process of NUSC. Task scheduling, data division, node communication, fault-tolerant recovery, and other parallelization tasks are managed by AutoNUSC. For the performance tuning issues of NUSC, this paper implements performance optimization strategies for AutoNUSC, including vectorization, memory access optimization, data reuse, etc. The experiments show that AutoNUSC effectively reduces the workload of users in developing NUSC applications in heterogeneous computing environments. Performance acceleration of up to 339 times is achieved within a single node compared to the serial program. AutoNUSC can efficiently perform task scheduling and fault-tolerant recovery across multiple nodes, with desirable scalability and robustness.

关键词： Heterogeneous Computing NUSC programming model Runtime Framework

来源：评论

学校读者我要写书评

暂无评论

Two-Sided Stable Matching Decision-Making Method Considering Matching Intention under a Hesitant Fuzzy Environment

引用

Computer modeling in Engineering & Sciences 2023年第5期135卷 1603-1623页

作者： Qi Yue Zhibin Deng School of Management Shanghai University of Engineering ScienceShanghai201620China School of Information Management Jiangxi University of Finance and EconomicsNanchang330013China

In this paper,a stable two-sided matching(TSM)method considering the matching intention of agents under a hesitant fuzzy environment is *** method uses a hesitant fuzzy element(HFE)as its ***,the HFE preference matrix is transformed into the normalized HFE preference *** this basis,the distance and the projection of the normalized HFEs on positive and negative ideal solutions are ***,the normalized HFEs are transformed into agent *** the stable matching constraints,a multiobjective programming model with the objective of maximizing the satisfactions of two-sided agents is *** on the agent satisfaction matrix,the matching intention matrix of two-sided agents is *** to the agent satisfaction matrix and matching intention matrix,the comprehensive satisfaction matrix is set ***,the multiobjective programming model based on satisfactions is transformed into a multiobjective programming model based on comprehensive *** the G-S algorithm,the multiobjective programming model based on comprehensive satisfactions is solved,and then the best TSM scheme is ***,a terminal distribution example is used to verify the feasibility and effectiveness of the proposed method.

关键词： Two-sided matching stable matching hesitant fuzzy element matching intention programming model

来源：评论

学校读者我要写书评

暂无评论

Ordinal-cardinal consensus analysis for large-scale group decision making with uncertain self-confidence

引用

INFORMATION FUSION 2023年第1期93卷 344-362页

作者： Meng, Fanyong Zhao, Dengyu Tan, Chunqiao Li, Zijun Nanjing Univ Informat Sci & Technol Sch Management Sci & Engn Nanjing 210044 Peoples R China Nanjing Audit Univ Sch Business Nanjing 211815 Peoples R China Cent South Univ Sch Business Changsha 410083 Peoples R China Nanjing Univ Informat Sci & Technol 932 South Lushan Rd Changsha 410083 Peoples R China

Consensus analysis is necessary for large-scale group decision making (LSGDM) for ensuring reasonable decision results. This paper offers a new method for LSGDM with uncertain self-confidence that follows ordinal-cardinal consensus analysis. For this purpose, a new interval ranking method is first proposed to compare alternatives. Then, an improved ordinal clustering method on criteria is introduced using the deviation between individual ranking positions. In view of opinion deviation, uncertain self-confidence deviation, and ranking deviation, the weights of decision makers (DMs) are defined. Similarly, the weights of clusters are determined by further combining cluster cardinality. Further, an ordinal-cardinal consensus procedure is offered, which contains two algorithms: the first algorithm is about the ordinal consensus improvement in view of three aspects: ranking adjustment, opinion adjustment, and uncertain self-confidence;the second algorithm studies the cardinal consensus improvement under the ordinal consensus requirement, which also mainly contains three aspects: opinion adjustment, the number of adjusted judgments, and uncertain self-confidence. Moreover, a new algo-rithm for LSGDM is presented. Finally, an example is provided to check the feasibility and efficiency of the new method, and a comparison analysis is also made.

关键词： LSGDM Ordinal clustering method Ordinal -cardinal consensus programming model

来源：评论

学校读者我要写书评

暂无评论

AtomiS: Data-Centric Synchronization Made Practical

引用

PROCEEDINGS OF THE ACM ON programming LANGUAGES-PACMPL 2023年第OOPSLA期7卷 116–145页

作者： Paulino, Herve Matos, Ana Almeida Cederquist, Jan Giunti, Marco Matos, Joao Ravara, Antonio NOVA Univ Lisbon Dept Comp Sci NOVA LINCS Caparica Portugal NOVA Univ Lisbon FCT NOVA Caparica Portugal Inst Telecomunicacoes SQIG Aveiro Portugal Univ Lisbon Dept Comp Sci & Engn & IST Lisbon Portugal Univ Lisbon Dept Comp Sci IST Lisbon Portugal

Data-Centric Synchronization (DCS) shifts the reasoning about concurrency restrictions from control structures to data declaration. It is a high-level declarative approach that abstracts away from the actual concurrency control mechanism(s) in use. Despite its advantages, the practical use of DCS is hindered by the fact that it may require many annotations and/or multiple implementations of the same method to cope with differently qualified parameters. To overcome these limitations, in this paper we present AtomiS, a new DCS approach that requires only qualifying types of parameters and return values in interface definitions, and of fields in class definitions. The latter may also be abstracted away in type parameters, rendering class implementations virtually annotationfree. From this high level specification, a static analysis infers the atomicity constraints that are local to each method, considering valid only the method variants that are consistent with the specification, and performs code generation for all valid variants of each method. The generated code is then the target for automatic injection of concurrency control primitives that are responsible for ensuring the absence of data-races, atomicity-violations and deadlocks. We provide a Java implementation and showcase the applicability of AtomiS in real-life code. For the benchmarks analysed, AtomiS requires fewer annotations than the original number of regions requiring locks, as well as fewer annotations than Atomic Sets (a reference DCS proposal).

关键词： Concurrency Data-Centric programming model Inference and Synthesis

来源：评论

学校读者我要写书评

暂无评论

Quantum programming Without the Quantum Physics 22nd

Quantum Programming Without the Quantum Physics

引用

22nd Asian Symposium on programming Languages and Systems

作者： Inoue, Jun Natl Inst Adv Ind Sci & Technol 1-8-31 Midorigaoka Ikeda Osaka Japan

ISBN: (纸本)9789819789429;9789819789436

We propose a quantum programming paradigm where all data are familiar classical data, and the only non-classical element is a random number generator that can return results with negative probability. Currently, the vast majority of quantum programming languages instead work with quantum data types made up of qubits. The description of their behavior relies on heavy linear algebra and many interdependent concepts and intuitions from quantum physics, which takes dedicated study to understand. We demonstrate that the proposed view of quantum programming explains its central concepts and constraints in more accessible, computationally relevant terms. This is achieved by systematically reducing everything to the existence of that negative-probability random generator, avoiding mention of advanced physics. This makes quantum programming more accessible to programmers without a deep background in physics or linear algebra. The bulk of this paper is written with such an audience in mind. As a working vehicle, we lay out a simple quantum programming language under this paradigm, showing that not only can it express all quantum algorithms, it also naturally captures the semantics of measurement without ever mentioning qubits or collapse.

关键词： Quantum programming Probabilistic programming programming model

来源：评论

学校读者我要写书评

暂无评论

Extending the SYCL Joint Matrix for Binarized Neural Networks

Extending the SYCL Joint Matrix for Binarized Neural Network...

引用

1st International Conference on Smart Energy Systems and Artificial Intelligence (SESAI)

作者： Jin, Zheming Oak Ridge Natl Lab Oak Ridge TN 37830 USA

ISBN: (纸本)9798350364613;9798350364606

In contrast to the warp matrix-multiplication application-programming interface (WMMA) for tensor hardware programming in Compute Unified Device Architecture (CUDA), the SYCL joint matrix extension is an experimental feature for devices that contain a matrix hardware. While the WMMA defines 1-bit precision and bit arithmetic operations as an experimental feature for applications such as binarized neural networks (BNNs), the SYCL joint matrix extension lacks such support. This paper describes the extension to the SYCI. matrix interface to leverage bit-capability of the tensor hardware. Then, we share the experience of migrating BNNs from CUDA to SYCL. Filially, we evaluate the performance of BNNs in CUDA and SYCI, on graphics processing units. We hope that the results will be useful for improving portability of the SYCL programming model.

关键词： GPU Tensor hardware programming model BAN

来源：评论

学校读者我要写书评

暂无评论

Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs

Understanding Performance Portability of SYCL Kernels: A Cas...

引用

37th IEEE International Parallel and Distributed Processing Symposium (IPDPS)

作者： Jin, Zheming Vetter, Jeffrey S. Oak Ridge Natl Lab Oak Ridge TN 37830 USA

ISBN: (纸本)9798350311990

SYCL is a portable programming model. Toward the goal of a better understanding of performance portability of SYCL kernels on GPUs, we select a bioinformatics kernel for computing the all-pairs distance as a case study. After migrating the kernel from CUDA to HIP and SYCL, we evaluate the performance of the CUDA, HIP, and SYCL kernels on NVIDIA V100 and AMD MI210 GPUs. We analyze the GPU instructions from the kernels to explain performance gaps between SYCL and CUDA/HIP. We hope that the findings are valuable for improving performance portability of SYCL.

关键词： Heterogeneous computing Performance portability programming model

来源：评论

学校读者我要写书评

暂无评论

APPy: Annotated Parallelism for Python on GPUs 2024

APPy: Annotated Parallelism for Python on GPUs

引用

33rd ACM SIGPLAN International Conference on Compiler Construction (CC)

作者： Zhou, Tong Shirako, Jun Sarkar, Vivek Georgia Inst Technol Atlanta GA 30332 USA

ISBN: (纸本)9798400705076

GPUs are increasingly being used used to speed up Python applications in the scientific computing and machine learning domains. Currently, the two common approaches to leveraging GPU acceleration in Python are 1) create a custom native GPU kernel, and import it as a function that can be called from Python;2) use libraries such as CuPy, which provides pre-defined GPU-implementation-backed tensor operators. The first approach is very flexible but requires tremendous manual effort to create a correct and high performance GPU kernel. While the second approach dramatically improves productivity, it is limited in its generality, as many applications cannot be expressed purely using CuPy's predefined tensor operators. Additionally, redundant memory access can often occur between adjacent tensor operators due to the materialization of intermediate results. In this work, we present APPy (Annotated Parallelism for Python), which enables users to parallelize generic Python loops and tensor expressions for execution on GPUs by adding simple compiler directives (annotations) to Python code. Empirical evaluation on 20 scientific computing kernels from the literature on a server with anAMDRyzen 7 5800X 8-Core CPU and an NVIDIA RTX 3090 GPU demonstrates that with simple pragmas APPy is able to generate more efficient GPU code and achieves significant geometric mean speedup relative to CuPy (30x on average), and to three state-of-the-art Python compilers, Numba (8.3x on average), DaCe-GPU (3.1x on average) and JAX-GPU (18.8x on average).

关键词： GPUs Python compilers programming model code generation

来源：评论

学校读者我要写书评

暂无评论

Improving the accessibility of NUMA-aware C plus plus application development based on the PGASUS framework

引用

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE 2023年第14期35卷 p1-24页

作者： Plauth, Max Eberhardt, Felix Grapentin, Andreas Polze, Andreas Univ Potsdam Hasso Plattner Inst Digital Engn Potsdam Germany

Certain workloads such as in-memory databases are inherently hard to scale-out and rely on cache-coherent scale-up non-uniform memory access (NUMA) systems to keep up with the ever-increasing demand for compute resources. However, many parallel programming frameworks such as OpenMP do not make efficient use of large scale-up NUMA systems as they do not consider data locality sufficiently. In this work, we present PGASUS, a C++ framework for NUMA-aware application development that provides integrated facilities for NUMA-aware task parallelism and data placement. The framework is based on an extensive review of parallel programming languages and frameworks to incorporate the best practices of the field. In a comprehensive evaluation, we demonstrate that PGASUS provides average performance improvements of 1.56x and peak performance improvements of up to 4.67xacross a wide range of workloads.

关键词： non-uniform memory access programming model scale-up computing

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Management of Extreme-Scale Task-Based Applications 29th

Hierarchical Management of Extreme-Scale Task-Based Applicat...

引用

29th International Conference on Parallel and Distributed Computing (Euro-Par)

作者： Lordan, Francesc Puigdemunt, Gabriel Verges, Pere Conejero, Javier Ejarque, Jorge Badia, Rosa M. Barcelona Supercomp Ctr Barcelona Spain Univ Calif Irvine Irvine CA USA

ISBN: (纸本)9783031396977;9783031396984

The scale and heterogeneity of exascale systems increment the complexity of programming applications exploiting them. Task-based approaches with support for nested tasks are a good-fitting model for them because of the flexibility lying in the task concept. Resembling the hierarchical organization of the hardware, this paper proposes establishing a hierarchy in the application workflow for mapping coarse-grain tasks to the broader hardware components and finer-grain tasks to the lowest levels of the resource hierarchy to benefit from lower-latency and higher-bandwidth communications and exploiting locality. Building on a proposed mechanism to encapsulate within the task the management of its finer-grain parallelism, the paper presents a hierarchical peer-to-peer engine orchestrating the execution of workflow hierarchies with fully-decentralized management. The tests conducted on the MareNostrum 4 supercomputer using a prototype implementation prove the validity of the proposal supporting the execution of up to 707,653 tasks using 2,400 cores and achieving speedups of up to 106 times faster than executions of a single workflow and centralized management.

关键词： distributed systems exascale task-based programming model workflow hierarchy runtime system peer-to-peer decentralized management

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：