检索结果-内蒙古大学图书馆

27th acm symposium on Operating Systems principles (SOSP)

作者： Yang, Ke Zhang, MingXing Chen, Kang Ma, Xiaosong Bai, Yang Jiang, Yong Tsinghua Univ Beijing Natl Res Ctr Informat Sci & Technol BNRis Dept Comp Sci & Technol Beijing Peoples R China Tsinghua Univ Res Inst Shenzhen Peoples R China Sangfor Technol Inc Shenzhen Peoples R China Hamad Bin Khalifa Univ Qatar Comp Res Inst Ar Rayyan Qatar 4Paradigm Co Ltd Beijing Peoples R China

ISBN: (纸本)9781450368735

Random walk on graphs has recently gained immense popularity as a tool for graph data analytics and machine learning. Currently, random walk algorithms are developed as individual implementations and suffer significant performance and scalability problems, especially with the dynamic nature of sophisticated walk strategies. We present KnightKing, the first general-purpose, distributed graph random walk engine. To address the unique interaction between a static graph and many dynamic walkers, it adopts an intuitive walker-centric computation model. The corresponding programming model allows users to easily specify existing or new random walk algorithms, facilitated by a new unified edge transition probability definition that applies across popular known algorithms. With KnightKing, these diverse algorithms benefit from its common distributed random walk execution engine, centered around an innovative rejection-based sampling mechanism that dramatically reduces the cost of higher-order random walk algorithms. Our evaluation confirms that KnightKing brings up to 4 orders of magnitude improvement in executing algorithms that currently can only be afforded with approximation solutions on large graphs.

关键词： graph computing random walk rejection sampling

来源：评论

学校读者我要写书评

暂无评论

CrashTuner: Detecting Crash-Recovery Bugs in Cloud Systems via Meta-Info Analysis 19

CrashTuner: Detecting Crash-Recovery Bugs in Cloud Systems v...

引用

27th acm symposium on Operating Systems principles (SOSP)

作者： Lu, Jie Liu, Chen Li, Lian Feng, Xiaobing Tan, Feng Yang, Jun You, Liang Univ Chinese Acad Sci CAS ICT SKL Comp Architecture Beijing Peoples R China Alibaba Grp Hangzhou Peoples R China TianQi Soft Inc Chengdu Peoples R China

ISBN: (纸本)9781450368735

Crash-recovery bugs (bugs in crash-recovery-related mechanisms) are among the most severe bugs in cloud systems and can easily cause system failures. It is notoriously difficult to detect crash-recovery bugs since these bugs can only be exposed when nodes crash under special timing conditions. This paper presents CrashTuner, a novel fault-injection testing approach to combat crash-recovery bugs. The novelty of CrashTuner lies in how we identify fault-injection points (crash points) that are likely to expose errors. We observe that if a node crashes while accessing meta-info variables, i.e., variables referencing high-level system state information (e.g., an instance of node or task), it often triggers crash-recovery bugs. Hence, we identify crash points by automatically inferring meta-info variables via a log-based static program analysis. Our approach is automatic and no manual specification is required. We have applied CrashTuner to five representative distributed systems: Hadoop2/Yarn, HBase, HDFS, ZooKeeper, and Cassandra. CrashTuner can finish testing each system in 17.39 hours, and reports 21 new bugs that have never been found before. All new bugs are confirmed by the original developers and 16 of them have already been fixed (14 with our patches). These new bugs can cause severe damages such as cluster down or start-up failures.

关键词： Crash Recovery Bugs Fault Tolerance distributed Systems Bug Detection Fault Injection Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Relays: Towards a link layer for robust and secure fog computing 37

Relays: Towards a link layer for robust and secure fog compu...

引用

2018 Workshop on Theory and Practice for Integrated Cloud, Fog and Edge computing Paradigms, TOPIC 2018, in conjunction with the 37th acm symposium on principles of distributed computing, PODC 2018

作者： Scheideler, Christian Paderborn University Paderborn Germany

In my presentation I will propose a new link-layer model for distributed computing based on so-called relays that promises to be useful for the design of robust and secure distributed systems based on overlay networks... 详细信息

ISBN: (纸本)9781450357760

关键词： Overlay networks

来源：评论

学校读者我要写书评

暂无评论

proceedings of the annual acm symposium on principles of distributed computing

Proceedings of the Annual ACM Symposium on Principles of Dis...

引用

2018 Workshop on Theory and Practice for Integrated Cloud, Fog and Edge computing Paradigms, TOPIC 2018, in conjunction with the 37th acm symposium on principles of distributed computing, PODC 2018

ISBN: (纸本)9781450357760

The proceedings contain 7 papers. The topics discussed include: Relays: towards a link layer for robust and secure fog computing;distributing computations in fog architectures;scheduling at the edge for assisting cloud real-time systems;enabling exclusive shared access to cloud of things resources;digital epidemiology and beyond;a novel NFV schedule optimization approach with sensitivity to packets dropping positions;and GoEdge: a scalable and stateless local breakout method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Turn of the carousel - what does edge computing change for distributed applications? 37

Turn of the carousel - what does edge computing change for d...

引用

2018 Workshop on Advanced Tools, Programming Languages, and PLatforms for Implementing and Evaluating Algorithms for distributed Systems, ApPLIED 2018, co-located with the 37th acm symposium on the principles of distributed computing, PODC 2018

作者： Venugopal, Srikumar Gkoufas, Yiannis Gazzetti, Michele Katrinis, Kostas IBM Research Ireland

ISBN: (纸本)9781450357753

Over the years, developments such as cloud computing, Internet of Things, and now edge and fog computing, have probably caused paradigm fatigue among practitioners. The question arises whether adopting a specific paradigm has a fundamental effect on the development and deployment of applications. This talk will examine this question in the context of edge computing, through the lens of developing and deploying an visual inference application. © 2018 Copyright held by the owner/author(s).

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

proceedings of the annual acm symposium on principles of distributed computing

Proceedings of the Annual ACM Symposium on Principles of Dis...

引用

ISBN: (纸本)9781450357753

The proceedings contain 9 papers. The topics discussed include: saying what you mean;towards reproducible evaluation of large-scale distributed systems;turn of the carousel – what does edge computing change for distributed applications?: research statement;towards a more reliable store-and-forward protocol for mobile text messages;logical clocks are not fair: what is fair? a case study of high-level language and optimization;an analysis of quorum-based abstractions: a case study using gorums to implement raft;data distribution method for fast giga-scale hologram generation on a multi-GPU cluster;language semantics driven design and formal analysis for distributed cyber-physical systems;and Partisan: enabling real-world protocol evaluation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Partisan: Enabling real-world protocol evaluation 37

Partisan: Enabling real-world protocol evaluation

引用

作者： Meiklejohn, Christopher S. Université Catholique de Louvain Louvain-la-Neuve Belgium

ISBN: (纸本)9781450357753

We present the design and implementation of Partisan, an Erlang library for enabling real-world experiments of distributed protocols and applications. Partisan is a "batteries-included" library facilitating internode communication in Erlang, runtime selection of cluster topology, pluggable layers that provide additional functionality such as causal delivery and reliable message delivery, and a mechanism for performing deterministic fault injection. Partisan has been used in the evaluation of one research prototype, two real-world applications, and has seen industry adoption in the Erlang and Elixir communities. © 2018 Copyright held by the owner/author(s).

关键词： distributed computer systems

来源：评论

学校读者我要写书评

暂无评论

proceedings of the annual acm symposium on principles of distributed computing

Proceedings of the Annual ACM Symposium on Principles of Dis...

引用

37th acm SIGACT-SIGOPS symposium on principles of distributed computing, PODC 2018

ISBN: (纸本)9781450357951

The proceedings contain 62 papers. The topics discussed include: nesting-safe recoverable linearizability: modular constructions for non-volatile memory;deterministic abortable mutual exclusion with sublogarithmic adaptive RMR complexity;brief announcement: persistent multi-word compare-and-swap;separating lock-freedom from wait-freedom;passing messages while sharing memory;revisionist simulations: a new approach to proving space lower bounds;on the classification of deterministic objects via set agreement power;leveraging indirect signaling for topology inference and fast broadcast;broadcast in radio networks, time vs. energy tradeoffs;round- and message-optimal distributed graph algorithms;improved massively parallel computation algorithms for mismatching, and vertex cover;distributed approximation of minimum k-edge-connected spanning subgraphs;brief announcement: distributed minimum vertex coloring and maximum independent set in chordal graphs;fair leader election for rational agents in asynchronous rings and networks;leader election in well-connected graphs;almost-surely terminating asynchronous byzantine agreement revisited;sublinear message bounds for randomized agreement;tight bounds for asymptotic and approximate consensus;property testing of planarity in the congest model;and locking timestamps versus locking objects.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Brief Tutorial on distributed and Concurrent Machine Learning 18

A Brief Tutorial on Distributed and Concurrent Machine Learn...

引用

37th acm SIGACT-SIGOPS symposium on principles of distributed computing (PODC)

作者： Alistarh, Dan IST Austria Klosterneuburg Austria

ISBN: (纸本)9781450357951

The area of machine learning has made considerable progress over the past decade, enabled by the widespread availability of large datasets, as well as by improved algorithms and models. Given the large computational demands of machine learning workloads, parallelism, implemented either through single-node concurrency or through multi-node distribution, has been a third key ingredient to advances in machine learning. The goal of this tutorial is to provide the audience with an overview of standard distribution techniques in machine learning, with an eye towards the intriguing trade-offs between synchronization and communication costs of distributed machine learning algorithms, on the one hand, and their convergence, on the other. The tutorial will focus on parallelization strategies for the fundamental stochastic gradient descent (SGD) algorithm, which is a key tool when training machine learning models, from classical instances such as linear regression, to state-of-the-art neural network architectures. The tutorial will describe the guarantees provided by this algorithm in the sequential case, and then move on to cover both shared-memory and message-passing parallelization strategies, together with the guarantees they provide, and corresponding trade-offs. The presentation will conclude with a broad overview of ongoing research in distributed and concurrent machine learning. The tutorial will assume no prior knowledge beyond familiarity with basic concepts in algebra and analysis.

关键词： Machine Learning distributed Systems Shared Memory Message Passing Stochastic Gradient Descent

来源：评论

学校读者我要写书评

暂无评论

An algorithmic approach to the asynchronous computability theorem

引用

Journal of Applied and Computational Topology 2018年第3-4期1卷 451-474页

作者： Saraph, Vikram Herlihy, Maurice Gafni, Eli Brown University Providence RI United States UCLA Los Angeles CA United States

The asynchronous computability theorem (ACT) uses concepts from combinatorial topology to characterize which tasks have wait-free solutions in read–write memory. A task can be expressed as a relation between two chromatic simplicial complexes. The theorem states that a task has a protocol (algorithm) if and only if there is a certain color-preserving simplicial map compatible with that relation. The original proof of the ACT, given by Herlihy and Shavit (proceedings of the 25th annual acm symposium on theory of computing, pp 111–120, 1993;J acm 46(6):858–923, 1999) relied on an involved geometric argument. Borowsky and Gafni (proceedings of the 16th annual acm symposium on principles of distributed computing, pp 189–198, 1997) later proposed an alternative proof based on a distributed algorithmic, termed the “convergence algorithm”. However the description of this algorithm was incomplete, and presented without proof. In this paper, we give the first complete description, along with a proof of correctness. © 2018, Springer International Publishing AG, part of Springer Nature.

关键词： Asynchronous computability Combinatorial algebraic topology distributed computing Link-connectivity

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：