检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

9,398 篇 会议
244 篇 期刊文献
14 册 图书

馆藏范围

9,656 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

5,292 篇 工学
- 5,080 篇 计算机科学与技术...
- 2,404 篇 软件工程
- 875 篇 电气工程
- 809 篇 信息与通信工程
- 332 篇 控制科学与工程
- 325 篇 电子科学与技术（可...
- 205 篇 网络空间安全
- 60 篇 生物医学工程（可授...
- 52 篇 生物工程
- 50 篇 动力工程及工程热...
- 48 篇 机械工程
- 43 篇 建筑学
- 39 篇 仪器科学与技术
- 36 篇 安全科学与工程
- 35 篇 测绘科学与技术
- 33 篇 光学工程
- 31 篇 土木工程
- 30 篇 化学工程与技术
1,215 篇 理学
- 1,022 篇 数学
- 137 篇 统计学（可授理学、...
- 105 篇 系统科学
- 88 篇 物理学
- 66 篇 生物学
- 41 篇 化学
- 36 篇 地球物理学
1,009 篇 管理学
- 847 篇 管理科学与工程(可...
- 312 篇 工商管理
- 199 篇 图书情报与档案管...
64 篇 经济学
- 63 篇 应用经济学
33 篇 法学
- 32 篇 社会学
22 篇 医学
19 篇 农学
12 篇 教育学
3 篇 文学
1 篇 军事学

主题

1,183 篇 parallel process...
929 篇 distributed proc...
684 篇 concurrent compu...
673 篇 computational mo...
664 篇 computer archite...
620 篇 computer science
607 篇 distributed comp...
506 篇 hardware
455 篇 application soft...
435 篇 graphics process...
434 篇 conferences
391 篇 parallel algorit...
383 篇 parallel program...
378 篇 algorithm design...
377 篇 runtime
331 篇 graphics process...
323 篇 scalability
322 篇 parallel process...
315 篇 processor schedu...
297 篇 libraries

机构

28 篇 ibm thomas j. wa...
25 篇 college of compu...
24 篇 school of comput...
23 篇 oak ridge nation...
23 篇 oak ridge natl l...
21 篇 oak ridge nation...
18 篇 georgia inst tec...
18 篇 pacific northwes...
18 篇 argonne national...
18 篇 pacific northwes...
17 篇 department of co...
17 篇 virginia tech de...
16 篇 department of co...
14 篇 school of comput...
14 篇 tsinghua univers...
14 篇 iit dept comp sc...
14 篇 sandia national ...
14 篇 georgia institut...
13 篇 lawrence berkele...
13 篇 argonne natl lab...

作者

37 篇 jack dongarra
26 篇 dongarra jack
24 篇 bader david a.
22 篇 nakano koji
17 篇 feng wu-chun
16 篇 ito yasuaki
16 篇 zomaya albert y.
16 篇 schulz martin
16 篇 azad ariful
16 篇 talbi el-ghazali
16 篇 david a. bader
15 篇 hoefler torsten
15 篇 p. sadayappan
14 篇 gagan agrawal
14 篇 prasad sushil k.
13 篇 wang guojun
13 篇 kishore kothapal...
13 篇 kalyanaraman ana...
13 篇 rajamanickam siv...
13 篇 marco d. santamb...

语言

9,639 篇 英文
14 篇 中文
3 篇 其他

检索条件"任意字段=IEEE International Parallel and Distributed Processing Symposium Workshops"

共 9656 条记录，以下是121-130 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

CuGraph C++ primitives: Vertex/edge-centric building blocks for parallel graph computing

CuGraph C++ primitives: Vertex/edge-centric building blocks ...

引用

2023 ieee international parallel and distributed processing symposium workshops, IPDPSW 2023

作者： Kang, Seunghwa Hastings, Chuck Eaton, Joe Rees, Brad NVIDIA United States

ISBN: (纸本)9798350311990

Software development of high-performance graph algorithms is difficult on modern parallel computers. To simplify this task, we have designed and implemented a collection of C++ graph primitives, basic building blocks, within cuGraph to assist graph analytics software developers on parallel computers, ranging from desktops to large clusters. This graph primitives API provides a vertex/edge-centric C++ Standard Template Library (STL)-like interface, allowing users to pick a primitive algorithm, and specify desired operations on vertices and edges and how to reduce the output of such operations through C++ functors. The API implementation is responsible for executing these functors on the underlying hardware. In this case, the graph primitives are implemented to run on NVIDIA GPU systems, from a single-GPU to multi-GPUs in a distributed cluster. RAPIDS cuGraph is NVIDIA's graph analytics solution for data scientists and software integrators. Algorithms in cuGraph are either implemented using the cuGraph C++ primitives API or being migrated over to using the primitives API. The Louvain and PageRank algorithms have been tested on clusters with over 1000 GPUs. © 2023 ieee.

关键词： C++ (programming language)

来源：评论

学校读者我要写书评

暂无评论

MDLoader: A Hybrid Model-driven Data Loader for distributed Deep Neural Networks Training

MDLoader: A Hybrid Model-driven Data Loader for Distributed ...

引用

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

作者： Bae, Jonghyun Choi, Jong Youl Pasini, Massimiliano Lupo Mehta, Kshitij Ibrahim, Khaled Z. Lawrence Berkeley Natl Lab Berkeley CA 94720 USA Oak Ridge Natl Lab Oak Ridge TN USA

In this work, we propose MDLoader, a hybrid in-memory data loader for distributed deep neural networks. MDLoader introduces a model-driven performance estimator to automatically switch between one-sided and collective... 详细信息

ISBN: (纸本)9798350364613;9798350364606

关键词： One-sided communication Collective communication Graph Neural Network Performance estimator

来源：评论

学校读者我要写书评

暂无评论

An Optimized Error-controlled MPI Collective Framework Integrated with Lossy Compression 38

An Optimized Error-controlled MPI Collective Framework Integ...

引用

international parallel and distributed processing symposium (IPDPS)

作者： Huang, Jiajun Di, Sheng Yu, Xiaodong Zhai, Yujia Zhang, Zhaorui Liu, Jinyang Lu, Xiaoyi Raffenetti, Ken Zhou, Hui Zhao, Kai Chen, Zizhong Cappello, Franck Guo, Yanfei Thakur, Rajeev Univ Calif Riverside Riverside CA 92521 USA Argonne Natl Lab Argonne IL USA Stevens Inst Technol Hoboken NJ USA Hong Kong Polytech Univ Hong Kong Peoples R China Univ Calif Merced Merced CA USA Florida State Univ Tallahassee FL USA

ISBN: (纸本)9798350387117;9798350387124

With the ever-increasing computing power of supercomputers and the growing scale of scientific applications, the efficiency of MPI collective communications turns out to be a critical bottleneck in large-scale distributed and parallel processing. The large message size in MPI collectives is particularly concerning because it can significantly degrade the overall parallel performance. To address this issue, prior research simply applies the off-the-shelf fix-rate lossy compressors in the MPI collectives, leading to suboptimal performance, limited generalizability, and unbounded errors. In this paper, we propose a novel solution, called C-Coll, which leverages error-bounded lossy compression to significantly reduce the message size, resulting in a substantial reduction in communication cost. The key contributions are three-fold. (1) We develop two general, optimized lossy-compression-based frameworks for both types of MPI collectives (collective data movement as well as collective computation), based on their particular characteristics. Our framework not only reduces communication cost but also preserves data accuracy. (2) We customize SZx, an ultra-fast error-bounded lossy compressor, to meet the specific needs of collective communication. (3) We integrate C-Coll into multiple collectives, such as MPI Allreduce, MPI Scatter, and MPI Bcast, and perform a comprehensive evaluation based on real-world scientific datasets. Experiments show that our solution outperforms the original MPI collectives as well as multiple baselines and related efforts by 1.8-2.7x.

关键词： Lossy Compression MPI Collective distributed Systems

来源：评论

学校读者我要写书评

暂无评论

Reservoir Echo State Network for Classification of Multivariate Time Series 30

Reservoir Echo State Network for Classification of Multivari...

引用

30th international Conference on High Performance Computing, Data, and Analytics (HiPC)

作者： Purkayastha, Basab Bijoy Barma, Shovan Indian Inst Technol Guwahati Dept Phys Gauhati India Indian Inst Informat Technol Guwahati Dept ECE Gauhati India

ISBN: (纸本)9798350383782;9798350383799

Multivariate time series (MTS) classification has been tackled using various methods, including Reservoir Computing (RC), which generates efficient vectorized representations like reservoir state (RS). RS shines when handling extensive classes or training sets but demands longer processing and substantial memory. Addressing this, in this study we present the parallel Reservoir Echo State Network (PR-ESN), an optimized parallel training and evaluation algorithm rooted in the ESN principle. It leverages both CPU-shared memory and parallel distributed memory architecture to efficiently capture reservoir state's optimal model space representation, addressing computational challenges in MTS analysis. Distinguishing itself from previous works, PR-ESN combines distributed parallel processing at the network level and shared memory multiprocessing at the node level. This results in reduced memory requirements and faster processing, making it a significant contribution to the field. Key features include PR-ESN's distributed training and evaluation, shared memory parallelization, and MSR concatenation for comprehensive analysis of distributed model space representations. Testing on real-world MTS and benchmark ECG data proves PR-ESN-based classifiers achieve superior accuracy and faster processing times with optimal memory usage. Testing on real-world MTS and benchmark ECG data proves PR-ESN-based classifiers achieve superior accuracy and faster processing times with optimal memory usage.

关键词： Shared Memory parallelism distributed Memory parallelism Echo State Network Reservoir Computing parallel Computing Multivariate Time Series Classification

来源：评论

学校读者我要写书评

暂无评论

Accelerating Packet processing in Container Overlay Networks via Packet-level parallelism 37

Accelerating Packet Processing in Container Overlay Networks...

引用

37th ieee international parallel and distributed processing symposium (IPDPS)

作者： Lei, Jiaxin Munikar, Manish Lu, Hui Jia, Rao Binghamton Univ Binghamton NY 13902 USA Univ Texas Arlington Arlington TX USA

ISBN: (纸本)9798350337662

Overlay networks serve as the de facto network virtualization technique for providing connectivity among distributed containers. Despite the flexibility in building customized private container networks, overlay networks incur significant performance loss compared to physical networks (i.e., the native). The culprit lies in the inclusion of multiple network processing stages in overlay networks, which prolongs the network processing path and overloads CPU cores. In this paper, we propose MFLOW, a novel packet steering approach to parallelize the in-kernel data path of network flows. MFLOW exploits packet-level parallelism in the kernel network stack by splitting the packets of the same flow into multiple micro-flows, which can be processed in parallel on multiple cores. MFLOW devises new, generic mechanisms for flow splitting while preserving in-order packet delivery with little overhead. Our evaluation with both micro-benchmarks and real-world applications demonstrates the effectiveness of MFLOW, with significantly improved performance - e.g., by 81% in TCP throughput and 139% in UDP compared to vanilla overlay networks. MFLOW even achieved higher TCP throughput than the native (e.g., 29.8 vs. 26.6 Gbps).

关键词： Packet processing Kernel Network Stack Container Overlay Networks

来源：评论

学校读者我要写书评

暂无评论

Exploiting Inter-Layer Expert Affinity for Accelerating Mixture-of-Experts Model Inference 38

Exploiting Inter-Layer Expert Affinity for Accelerating Mixt...

引用

international parallel and distributed processing symposium (IPDPS)

作者： Yao, Jinghan Anthony, Quentin Shafi, Aamir Subramoni, Hari Panda, Dhabaleswar K. (DK) Ohio State Univ Columbus Dept Comp Sci & Engn Columbus OH 43210 USA

ISBN: (纸本)9798350387117;9798350387124

In the realm of large language models (LLMs) like the Generative Pre-trained Transformer (GPT), the Mixture of Experts (MoE) paradigm has emerged as a powerful technique for enhancing model expressiveness and accuracy. However, the deployment of GPT MoE models for parallel inference on distributed systems presents significant challenges, primarily due to the extensive Alltoall communication required for expert routing and aggregation. This communication bottleneck exacerbates the already complex computational landscape, hindering the efficient utilization of high-performance computing resources. In this paper, we propose a lightweight optimization technique called ExFlow, to largely accelerate the inference of these MoE models. We take a new perspective on alleviating the communication overhead by exploiting the inter-layer expert affinity. Unlike previous methods, our solution can be directly applied to pre-trained MoE models without any fine-tuning or accuracy degradation. By proposing a context-coherent expert parallelism on distributed systems, our ExFlow design only uses one Alltoall communication to deliver the same functionality while previous methods all require two Alltoalls. By carefully examining the conditional probability in tokens' routing across multiple layers, we proved that pre-trained GPT MoE models implicitly exhibit a strong inter-layer expert affinity. We then design an efficient integer programming model to precisely capture such features and show that by properly placing the experts on corresponding GPUs, we can reduce up to 67% of tokens' cross-GPU routing latency on various hardware configurations and topologies. Our solution beats the cutting-edge Deepspeed-MoE in GPT MoE models with experts from 8 to 64, with up to 2.2x improvement in inference throughput. To the best of our knowledge, this is the first work in leveraging inter-layer expert affinity to accelerate the inference of GPT MoE models. We further provide a detailed study of how

关键词： Mixture of experts parallel inference Collective communication Generative models distributed system

来源：评论

学校读者我要写书评

暂无评论

parallel Minimum Spanning Tree Algorithms via Lattice Linear Predicate Detection 36

Parallel Minimum Spanning Tree Algorithms via Lattice Linear...

引用

36th ieee international parallel and distributed processing symposium (ieee IPDPS)

作者： Alves, David R. Garg, Vijay K. Univ Texas Austin Dept Elect & Comp Engn Austin TX 78712 USA

ISBN: (纸本)9781665497473

We show that the problem of computing the minimum spanning tree can be formulated as special case of detecting Lattice Linear Predicate (LLP). In general, formulating problems as LLP presents two main advantages: 1) Different problems are formulated under a single, general framework, which defines the problem in terms of simple local predicates that must hold for the all the elements of a lattice, making the problem (and the solution) compact and easy to understand. 2) improvements on one set of problems can be transferable to other sets of problems;3) since the problems are stated as a set of local predicates, which can be often tested with little or no synchronization it is often the case that new opportunities for parallelism present themselves. In this paper we introduce two parallel algorithms LLP-Prim and LLP-Boruvka that improve on the non-LLP counterparts in several ways. LLP-Prim reduces the number of heap operations required by Prim by allowing edges to be selected without entering the heap thus allowing for parallelism. LLP-Boruvka improves on Boruvka by reducing synchronization and thus once more improving parallelism opportunities. Our experimental evaluation shows that LLP-Prim is faster than Prim's algorithm in both single threaded and multithreaded scenarios and that it provides a good tradeoff between parallelism and efficiency at low core counts. For higher core count scenarios we show how LLP-Boruvka improves on an efficient implementation of a parallel version of Boruvka.

关键词： Minimum Spanning Tree parallel Algorithms

来源：评论

学校读者我要写书评

暂无评论

Balancing Intra-Die and Inter-Die Placement Optimization in 2.5D FPGA Architectures

Balancing Intra-Die and Inter-Die Placement Optimization in ...

引用

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

作者： Raikar, Raveena Stroobandt, Dirk Univ Ghent Ghent Belgium

ISBN: (纸本)9798350364613;9798350364606

Interposer-based 2.5D FPGAs contain multiple dies interconnected through external wires, which introduce higher delay and longer wirelength compared to the interconnection network of a single die. Efficient CAD tools are needed for such architectures to strategically manage signal counts at the die boundary. In this study, we introduce a modular approach for parallel die-level placement, eliminating the need for a time-consuming global placement optimization step. Our methodology uses netlist partitioning and adds virtual anchor blocks to enable parallel die placement and facilitate inter-partition net optimization. Our method results in a multi-die placement tool 'LiquidMD'. Comparing our base multi-die placer with a state-of-the-art tool that fully globally optimizes placement shows substantial runtime improvement with improved placement quality. We also introduce a synchronization strategy for optimizing interpartition nets and assess how minimal synchronization benefits modular parallel placement. Our research results in a modular placement algorithm that strikes a favourable balance between runtime efficiency and placement quality.

关键词： 2.5D architectures Multi-die FPGA placement parallel placement Silicon Interposer

来源：评论

学校读者我要写书评

暂无评论

Teaching High-Performance Computing in Developing Countries: A Case Study in Mexican Universities 36

Teaching High-Performance Computing in Developing Countries:...

引用

36th ieee international parallel and distributed processing symposium (ieee IPDPS)

作者： Antonio Trejo-Sanchez, Joel Javier Hernandez-Lopez, Francisco Uh Zapata, Miguel Angel Luis Lopez-Martinez, Jose Fajardo-Delgado, Daniel Cesar Ramirez-Pacheco, Julio CONACYT Ctr Invest Matemat AC Dept Comp Sci Merida Mexico Univ Autonoma Yucatan Dept Comp Sci Merida Mexico Tecnol Nacl Mexico Dept Syst & Computat Ciudad Guzman Mexico Univ Quintana Roo Dept Engn Cancun Mexico

ISBN: (纸本)9781665497473

Teaching High-Performance Computing (HPC) to undergraduate programs represents a significant challenge in most universities in developing countries like Mexico. Deficiencies in the required infrastructure and equipment, inadequate curricula in computer engineering programs (and resistance to change them), students' lack of interest, motivation, or knowledge of this area are the main difficulties to overcome. The COVID-19 pandemic represents an additional challenge to these difficulties in teaching HPC in these programs. Despite the detriments, some strategies have been developed to incorporate the HPC concepts to Mexican students without necessarily modifying the traditional curricula. This paper presents a case study over four public universities in Mexico based on our experience as instructors. We also propose a course that introduces the HPC principles considering the heterogeneous background of the students in such universities. The results are about the number of students enrolling in related classes and participating in extra-curricular projects.

关键词： High-Performance Computing undergraduate program curricula parallel computing

来源：评论

学校读者我要写书评

暂无评论

Moving from Composable to Programmable 36

Moving from Composable to Programmable

引用

36th ieee international parallel and distributed processing symposium (ieee IPDPS)

作者： Chen, Zhongyi Renambot, Luc Long, Lance Brown, Maxine Johnson, Andrew E. Univ Illinois Elect Visualizat Lab Comp Sci Chicago IL 60607 USA

ISBN: (纸本)9781665497473

In today's Big Data era, data scientists require modern workflows to quickly analyze large-scale datasets using complex codes to maintain the rate of scientific progress. These scientists often rely on available campus resources or off-the-shelf computational systems for their applications. Unified infrastructure or over-provisioned servers can quickly become bottlenecks for specific tasks, wasting time and resources. Composable infrastructure helps solve these problems by providing users with new ways to increase resource utilization. Composable infrastructure disaggregates a computer's components - CPU, GPU (accelerators), storage and networking - into fluid pools of resources, but typically relies upon infrastructure engineers to architect individual machines. Infrastructure is either managed with specialized command-line utilities, user interfaces or specification files. These management models are cumbersome and difficult to incorporate into data-science workflows. We developed a high-level software API, Composastructure, which, when integrated into modern workflows, can be used by infrastructure engineers as well as data scientists to reorganize composable resources on demand. Composastructure enables infrastructures to be programmable, secure, persistent and reproducible. Our API composes machines, frees resources, supports multi-rack operations, and includes a Python module for Jupyter Notebooks.

关键词： distributed systems testbed implementation and deployment composable infrastructure deep learning visualization infrastructure as code

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共500页 << < 9 10 11 12 13 14 15 16 17 18 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：