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

4,316 篇 会议
151 篇 期刊文献
6 册 图书

馆藏范围

4,472 篇 电子文献
1 种 纸本馆藏

日期分布

学科分类号

3,283 篇 工学
- 3,129 篇 计算机科学与技术...
- 1,640 篇 软件工程
- 413 篇 信息与通信工程
- 374 篇 电气工程
- 224 篇 控制科学与工程
- 82 篇 电子科学与技术（可...
- 53 篇 网络空间安全
- 48 篇 生物工程
- 34 篇 机械工程
- 30 篇 建筑学
- 24 篇 动力工程及工程热...
- 22 篇 生物医学工程（可授...
- 20 篇 化学工程与技术
- 19 篇 环境科学与工程（可...
- 18 篇 仪器科学与技术
- 18 篇 农业工程
- 16 篇 材料科学与工程（可...
807 篇 理学
- 672 篇 数学
- 129 篇 统计学（可授理学、...
- 66 篇 系统科学
- 52 篇 生物学
- 43 篇 物理学
- 26 篇 化学
376 篇 管理学
- 225 篇 管理科学与工程(可...
- 176 篇 图书情报与档案管...
- 171 篇 工商管理
120 篇 教育学
- 120 篇 教育学
46 篇 法学
- 37 篇 社会学
45 篇 经济学
- 44 篇 应用经济学
17 篇 农学
- 17 篇 作物学
16 篇 医学
1 篇 文学
1 篇 军事学

主题

377 篇 distributed comp...
259 篇 distributed comp...
134 篇 distributed algo...
107 篇 grid computing
99 篇 cloud computing
93 篇 distributed syst...
91 篇 computational mo...
61 篇 concurrent compu...
61 篇 computer network...
55 篇 computer science
53 篇 fault tolerance
50 篇 distributed algo...
49 篇 application soft...
48 篇 algorithms
46 篇 parallel computi...
44 篇 computer archite...
44 篇 distributed data...
43 篇 message passing
42 篇 computer systems...
41 篇 high performance...

机构

47 篇 swiss fed inst t...
32 篇 technion haifa
30 篇 universidade do ...
27 篇 carnegie mellon ...
24 篇 mit cambridge ma...
22 篇 univ freiburg fr...
17 篇 vienna universit...
16 篇 univ michigan an...
15 篇 university of au...
14 篇 tel aviv univ te...
14 篇 natl univ singap...
13 篇 georgia inst tec...
12 篇 univ toronto on
12 篇 uas technikum vi...
12 篇 ben gurion univ ...
12 篇 cornell univ ith...
12 篇 aalto univ espoo
12 篇 univ calif berke...
11 篇 ecole polytech f...
11 篇 indiana univ blo...

作者

33 篇 kuhn fabian
31 篇 ghaffari mohsen
24 篇 elkin michael
22 篇 awerbuch baruch
22 篇 chen haibo
22 篇 pandurangan gopa...
21 篇 suomela jukka
20 篇 censor-hillel ke...
19 篇 peleg david
19 篇 fraigniaud pierr...
18 篇 chang yi-jun
18 篇 rui oliveira
17 篇 raynal michel
17 篇 wattenhofer roge...
16 篇 rajsbaum sergio
16 篇 guerraoui rachid
16 篇 kutten shay
16 篇 brandt sebastian
15 篇 karl m. goeschka
15 篇 patt-shamir boaz

语言

4,388 篇 英文
81 篇 其他
4 篇 中文

检索条件"任意字段=Proceedings of the ACM Symposium on Principles of Distributed Computing"

共 4473 条记录，以下是631-640 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：

AITurbo: Unified Compute Allocation for Partial Predictable Training in Commodity Clusters 21

AITurbo: Unified Compute Allocation for Partial Predictable ...

引用

30th International symposium on High-Performance Parallel and distributed computing (HPDC)

作者： Zhao, Laiping Li, Fangshu Qu, Wenyu Zhan, Kunlin Zhang, Qingman Tianjin Univ Coll Intelligence & Comp CIC Tianjin Key Lab Adv Networking TANKLAB Tianjin Peoples R China Tianjin Univ TANKLAB CIC Tianjin Peoples R China 58 Com Beijing Peoples R China UNICLOUD Tianjin Peoples R China

ISBN: (纸本)9781450382175

As the scale and complexity of deep learning models continues to grow, model training is becoming an expensive job and only a small number of well-financed organizations can afford. Are the resources in commodity clusters well utilized for training? or how much potential space are still there for further improving the training efficiency in commodity clusters? is an urgent question to answer. In this paper, we review the processing of distributed learning training (DDL) in commodity GPU clusters and find that the current resource utilization is not only low but also imbalanced. We observe two features that can be exploited for further improving the training efficiency: partial predictable training and unified CPU-GPU training. Based on the observations, we present AITurbo, a novel resource scheduler that treats predictable and unpredictable jobs separately, but allocates heterogeneous CPU-GPU resource in a unified way. For predictable jobs, AITurbo designs a predicting model to estimate their performance under various heterogeneous resource allocations. For unpredictable jobs, it schedules them following the least-attained-service-first manner. AITurbo further designs a Borda-count based multi-level feedback queue method to combine them together. AITurbo demonstrates that there is still significant space for improving the training efficiency in commodity clusters. We evaluate AITurbo using jobs from Tensorflow benchmarks, which are submitted following the real trace of three production systems. Experimental results show that, compared with the state-of-the-art, AITurbo can reduce the average job completion time of DDL jobs by 3x.

关键词： Scheduling distributed deep learning partial predictable training heterogeneous resource

来源：评论

学校读者我要写书评

暂无评论

Noisy Beeps 20

Noisy Beeps

引用

39th symposium on principles of distributed computing

作者： Efremenko, Klim Kol, Gillat Saxena, Raghuvansh R. Ben Gurion Univ Negev Beer Sheva Israel Princeton Univ Princeton NJ 08544 USA

ISBN: (纸本)9781450375825

We study the effect of noise on the n-party beeping model. In this model, in every round, each party may decide to either 'beep' or not. All parties hear a beep if and only if at least one party beeps. The beeping model is becoming increasingly popular, as it offers a very simple abstraction of wireless networks and is very well suited for studying biological phenomena. Still, the noise resilience of the beeping model is yet to be understood. Our main result is a lower bound, showing that making protocols in the beeping model resilient to noise may have a large performance overhead. Specifically, we give a protocol that works over the (noiseless) beeping model, and prove that any scheme that simulates this protocol over the beeping model with correlated stochastic noise will blow up the number of rounds by an O(logn) multiplicative factor. We complement this result by a matching upper bound, constructing a noise-resilient simulation scheme with O(logn) overhead for any noiseless beeping protocol.

关键词： Communication Complexity Beeping Models Lower Bounds Interactive Coding Error-Correcting Codes distributed Systems

来源：评论

学校读者我要写书评

暂无评论

Interaction Is Necessary for distributed Learning with Privacy or Communication Constraints 2020

Interaction Is Necessary for Distributed Learning with Priva...

引用

52nd Annual acm SIGACT symposium on Theory of computing (STOC)

作者： Dagan, Yuval Feldman, Vitaly MIT EECS 77 Massachusetts Ave Cambridge MA 02139 USA Google Res Mountain View CA USA Apple Cupertino CA USA

ISBN: (纸本)9781450369794

Local differential privacy (LDP) is a model where users send privatized data to an untrusted central server whose goal it to solve some data analysis task. In the non-interactive version of this model the protocol consists of a single round in which a server sends requests to all users then receives their responses. This version is deployed in industry due to its practical advantages and has attracted significant research interest. Our main result is an exponential lower bound on the number of samples necessary to solve the standard task of learning a large-margin linear separator in the non-interactive LDP model. Via a standard reduction this lower bound implies an exponential lower bound for stochastic convex optimization and specifically, for learning linear models with a convex, Lipschitz and smooth loss. These results answer the questions posed by Smith, Thakurta, and Upadhyay (IEEE symposium on Security and Privacy 2017) and Daniely and Feldman (NeurIPS 2019). Our lower bound relies on a new technique for constructing pairs of distributions with nearly matching moments but whose supports can be nearly separated by a large margin hyperplane. These lower bounds also hold in the model where communication from each user is limited and follow from a lower bound on learning using non-adaptive statistical queries.

关键词： Local Differential Privacy distributed Learning Interactive Protocol Communication-constrained Learning Statistical Query

来源：评论

学校读者我要写书评

暂无评论

King Louie: Reproducible Availability Benchmarking of Cloud-hosted DBMS 20

King Louie: Reproducible Availability Benchmarking of Cloud-...

引用

35th Annual acm symposium on Applied computing (SAC)

作者： Seybold, Daniel Wesner, Stefan Domaschka, Joerg Ulm Univ Inst Informat Resource Management Ulm Germany

ISBN: (纸本)9781450368667

Cloud resources have become a preferred operational model distributed Database Management Systems (DBMS) by offering the elasticity and virtually unlimited scalability, but increase the risk of failures with increasing cluster sizes. While distributed DBMS provide high-availability mechanisms, it is currently an open research question to what extent they are able to provide availability and performance guarantees in case of cloud resource failures. Especially as existing DBMS benchmarks do not consider availability. We present a comprehensive methodology for evaluating the availability of distributed DBMS in case of cloud resource failures. Based on this methodology, we introduce a novel framework that automates the full evaluation process, including the failure injection, and emphasizes reproducibility. The framework is validated by 16 diverse availability evaluations. The results show that distributed DBMS are not necessary available even if sufficient replicas are available and clients can experience significant downtimes.

关键词： availability NoSQL cloud distributed DBMS benchmark

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Load Balancing in distributed computing Environment: An Evolutionary computing Approach 20

Multi-objective Load Balancing in Distributed Computing Envi...

引用

35th Annual acm symposium on Applied computing (SAC)

作者： Kishor, Avadh Niyogi, Rajdeep Indian Inst Technol Dept Comp Sci & Engn Roorkee Uttar Pradesh India

ISBN: (纸本)9781450368667

In this paper, we consider the load balancing problem as a bi-objective problem which includes minimization of jobs' response time and utilization imbalance among servers. We consider both the objectives in an integrated manner and formulated as an optimization model. We cast into the game-theoretic setting. To derive the solution of the game, we propose a co-evolutionary computing framework, called CO-evolutionary framework based on Differential Evolution (CODE). Simulation results show that the CODE not only minimizes the jobs' response time but also significantly reduces the utilization imbalance among servers.

关键词： Load balancing Cloud environment Fair utilization Evolutionary computing Differential evolution

来源：评论

学校读者我要写书评

暂无评论

Swift: Expedited Failure Recovery for Large-Scale DNN Training 23

Swift: Expedited Failure Recovery for Large-Scale DNN Traini...

引用

proceedings of the 28th acm SIGPLAN Annual symposium on principles and Practice of Parallel Programming

作者： Yuchen Zhong Guangming Sheng Juncheng Liu Jinhui Yuan Chuan Wu The University of Hong Kong China OneFlow Inc. China

ISBN: (纸本)9798400700156

As the size of deep learning models gets larger and larger, training takes longer time and more resources, making fault tolerance critical. Existing state-of-the-art methods like Check-Freq and Elastic Horovod need to back up a copy of the model state in memory, which is costly for large models and leads to non-trivial overhead. This paper presents Swift, a novel failure recovery design for distributed deep neural network training that significantly reduces the failure recovery overhead without affecting training throughput and model accuracy. Instead of making an additional copy of the model state, Swift resolves the inconsistencies of the model state caused by the failure and exploits replicas of the model state in data parallelism for failure recovery. We propose a logging-based approach when replicas are unavailable, which records intermediate data and replays the computation to recover the lost state upon a failure. Evaluations show that Swift significantly reduces the failure recovery time and achieves similar or better training throughput during failure-free execution compared to state-of-the-art methods without degrading final model accuracy.

关键词： distributed DNN training failure resilience

来源：评论

学校读者我要写书评

暂无评论

Towards a distributed Planning of Decision Making Under Uncertainty for a Fleet of Robots 20

Towards a Distributed Planning of Decision Making Under Unce...

引用

35th Annual acm symposium on Applied computing (SAC)

作者： Jourdan, Paul Lozenguez, Guillaume Fabresse, Luc Bouraqadi, Noury INRAE Occitanie Toulouse Auzeville Haute Garonne France Univ Lille IMT Lille Douai Douai Nord France

ISBN: (纸本)9781450368667

Coordination is required in order to solve a multi robot navigation problem and allow an efficient and fast search of a solution while avoiding any possible collisions. Planning with a fleet of robots can rely on Multi-agent Markovian Decision Processes (MMDPs) model This assumes that it is possible to share the local perceptions of robots every time. However the computation of a distributed policy is not necessarily distributable between robots as with multiple path planning, where the movement of one robot depends on all the other's paths. The global search space would be of exponential size (in the number of robots) in most of multi-robot scenarios. distributed planning over the robots would allow each robot to plan its own policy while taking advantage of parallel computing. In this paper, this problem is addressed by presenting an approach consisting in starting from a simplified model which can be distributed, then by adding robots interactions constraints while maintaining the model distributable. The results of the experimentations with different configurations highlight some of the strength and limitations of the current approach.

关键词： Markov Decision Process Mobile Robotics Agent-Based Modeling Coordination

来源：评论

学校读者我要写书评

暂无评论

A distributed approach for autonomous cooperative transportation in a dynamic multi-robot environment 20

A distributed approach for autonomous cooperative transporta...

引用

35th Annual acm symposium on Applied computing (SAC)

作者： Nath, Amar Arun, A. R. Niyogi, Rajdeep IIT Roorkee Roorkee 247667 Uttar Pradesh India

ISBN: (纸本)9781450368667

This paper describes a distributed approach for autonomous cooperative transportation in a dynamic multi-robot environment. The proposed approach forms an optimal coalition at runtime for cooperative transportation and assigns a group of robots for the task. Explicit communication is used to acquire information as the robots do not have a global knowledge of the environment, i.e., no robot knows the location and state of another robot. For cooperative transportation, such information is essential as objects may arrive at any time and at any location. The proposed approach deals with on-demand missions, where the number of robots required to solve the problem is not known a priori. The applicability of the approach is demonstrated on a road clearance scenario in a realistic urban search and rescue simulation environment. The experimental results validate the correctness of the approach.

关键词： Multi-robot system coperative object transportation coalition formation distributed algorithm multi-robot coordination

来源：评论

学校读者我要写书评

暂无评论

distributed inference for multiple DNN models in IoT environments: poster 22

Distributed inference for multiple DNN models in IoT environ...

引用

proceedings of the Twenty-Third International symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile computing

作者： YoungHwan Jin HyungBin Park SuKyoung Lee Yonsei University Republic of Korea

ISBN: (纸本)9781450391658

Edge computing (EC) has been promising in providing support for Deep Neural Network (DNN) applications in IoT environments. However, resources in each edge are limited as response time increases. To reduce the latency, we propose to distribute the computation of multiple DNN models to nearby IoT devices. In particular, we propose a piece-wise multilevel partitioning and scheduling algorithm to improve the completion time of DNN inference.

关键词： IoT

来源：评论

学校读者我要写书评

暂无评论

Towards a Replication Service for Data-Intensive Fog Applications 20

Towards a Replication Service for Data-Intensive Fog Applica...

引用

35th Annual acm symposium on Applied computing (SAC)

作者： Hasenburg, Jonathan Grambow, Martin Bermbach, David TU Berlin Berlin Germany Einstein Ctr Digital Future Mobile Cloud Comp Res Berlin Germany

ISBN: (纸本)9781450368667

The combination of edge and cloud in the fog computing paradigm enables a new breed of data-intensive applications. These applications, however, have to face a number of fog-specific challenges which developers have to repetitively address for every single application. In this paper, we derive a set of requirements for a replication service that aims to simplify the development of data-intensive fog applications which are caused by the highly distributed and heterogeneous operation environment. Furthermore, we propose the design for such a service which addresses our requirements.

关键词： Fog computing Data Management Replication Service

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448页 << < 60 61 62 63 64 65 66 67 68 69 > >>

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：