检索结果-内蒙古大学图书馆

43rd International symposium on Reliable distributed Systems

作者： Qin, Kailun Gu, Dawu Shanghai Jiao Tong Univ Shanghai Peoples R China

ISBN: (纸本)9798331530044;9798331530037

Model Compilation as a Service (MCaaS) has emerged as critical Machine Learning (ML) supply chain infrastructure. It provides large-scale model optimization for heterogeneous hardware devices in an easy-to-use, cost-efficient and fault-tolerant manner. However, a variety of attacks targeting the ML model supply chain have been reported. Further compounded by the complexity of the cloud environments where MCaaS operates, increasing security concerns have been raised about the generated model binaries. In response, we present THEMIS - confidential MCaaS with privacy-preserving transparency. To help build trust in model binaries, we increase supply chain transparency by introducing property-based integrity that captures complex property evidence throughout the compilation pipeline. Meanwhile, we underscore privacy preservation through the concept of a model property tree, which represents the provenance and integrity of a model binary. It allows constrained sharing and efficient verification of binary properties. In THEMIS, we enforce confidentiality to all generated binaries by default, but with fine-grained control through the use of hybrid selective encryption based on attribute-based encryption. THEMIS is further secured within distributed Trusted Execution Environments to ensure confidential and reliable execution. We build THEMIS on top of a state-of-the-art ML compiler, with our evaluations demonstrating its practicality.

关键词： machine learning supply chain security property-based integrity merkle tree attribute-based encryption confidential computing

来源：评论

学校读者我要写书评

暂无评论

CLUES: Collusive Theft of Conditional Generative Adversarial Networks 43

CLUES: Collusive Theft of Conditional Generative Adversarial...

引用

43rd International symposium on Reliable distributed Systems

作者： Queyrut, Simon Schiavoni, Valerio Chen, Lydia Y. Felber, Pascal Birke, Robert Univ Neuchatel Inst Comp Sci IIUN Neuchatel Switzerland Univ Turin Turin Italy

ISBN: (纸本)9798331530044;9798331530037

Conditional Generative Adversarial Networks (cGANs) are increasingly popular web-based synthesis services accessed through a query API, e.g., cGANs generate a cat image based on a "cat" query. However, cGAN-based synthesizers can be stolen via adversaries' queries, i.e., model thieves. The prevailing adversarial assumption is that thieves act independently: they query the deployed cGAN (i.e., the victim), and train a stolen cGAN using the images obtained from the victim. A popular anti-theft defense consists in throttling down the number of queries from any given user. We consider a more realistic adversarial scenario: model thieves collude to query the victim, and then train the stolen cGAN. CLUES is a new collusive model stealing framework, enabling thieves to bypass throttle-based defenses and steal cGANs more efficiently than through individual efforts. Thieves collect queried images and train a stolen cGAN in a federated manner. We evaluate CLUES on three image datasets, e.g., MNIST, FashionMNIST and CelebA. We experimentally show the scalability of the proposed attack strategies against the number of thieves and the queried images, the impact of a classical noise-based defense, a passive watermarking defense and a JPEG-based countermeasure. Our evaluation shows that such a collusive stealing strategy gets close to 4 units of Frechet Inception Distance from a victim model. Our code is readily available to the research community: https://***/records/10224340.

关键词： cGAN distributed stealing model stealing stealing

来源：评论

学校读者我要写书评

暂无评论

Towards a Comprehensive Approach to Resource and Conflict Management in Cloud-Edge Settings 24

Towards a Comprehensive Approach to Resource and Conflict Ma...

引用

33rd International symposium on High-Performance Parallel and distributed computing (HPDC)

作者： Massa, Jacopo Univ Pisa Dept Comp Sci Pisa Italy CNR ISTI Pisa Italy

ISBN: (纸本)9798400704130

As the Internet of Things (IoT) gains popularity, the multitude of connected devices accessing computing facilities along the Cloud-IoT continuum has risen rapidly. This growth has prompted an increase in IoT-based applications and the data they process. These applications often carry specific Quality of Service (QoS) requirements, necessitating the research of strategic management for services and data, coupled with routes across the IoT and application components. Concurrently, our research explores the integration of Intent-Based Networking (IBN) principles to manage data flows and resolve conflicts among stakeholders sharing the same infrastructure. This dual-focus approach aims to support the effective placement and management of multi-service applications on Cloud-IoT resources, considering QoS, data characteristics, and resource conflict detection and resolution. This work outlines our primary objectives and shares findings from our ongoing research.

关键词： Cloud computing Intent-Based Networking Network resources allocation Constraint and logic languages

来源：评论

学校读者我要写书评

暂无评论

Keynote: Time is not a Healer: Before and After 24

Keynote: Time is not a Healer: Before and After

引用

proceedings of the 43rd acm symposium on principles of distributed computing

作者： Nicola Santoro School of Computer Science Carleton University Ottawa Canada

ISBN: (纸本)9798400706684

distributed computing has been concerned with the topics of faults and failures from its very beginning (before PODC was born). The consensus problem was at the forefront of the research efforts, and an extensive body of literature on the subject was developed very quickly (e.g., [5, 6, 11, 20]). The beautiful proof of the impossibility of consensus under asynchrony [7] put to rest the unsuccessful attempts to prove otherwise, and gave an even stronger impetus to the focus on synchronous *** paper Time is not a Healer [16], which we shall denote by {T¬H}, was concerned with the problem of agreement in synchronous systems where failures have mostly a transient and ubiquitous nature; that is, faults can occur anywhere in the system and, following a failure, normal functioning can resume after a finite (although unpredictable) time. These type of dynamic failures, however, could not be handled by the traditional failure models developed for the consensus and related *** the significance and novelty of the scientific and technical contributions of {T¬H}, including the model, the proof technique, and the unexpected results. Incidentally, while known for its results on omissions, {T¬H} contains also tight bounds on corruptions, additions, and the Byzantine combination of all. While known for its impossibility results, it has tight possibility results, which hold in situations for which the traditional models indicate impossibility. Over the decades, triggered by these results, an extensive literature has been developed (e.g., [3, 4, 8, 9, 13, 15, 19, 21]).The relatively recent developments in communication technology and advent of new virtual systems (e.g., wireless mobile networks, social networks) have driven the research in distributed computing to investigate problems (including consensus) in dynamic networks (e.g., [2, 10, 12]); this has inevitably brought to light the immediate connection between such networks and the ones subject to dynamic omiss

关键词： theory of distributed computation synchronous systems dynamic faults agreement dynamic networks

来源：评论

学校读者我要写书评

暂无评论

Emotional Design and User Study on Guidance System For MR distributed Collaboration 10

Emotional Design and User Study on Guidance System For MR Di...

引用

Joint Meeting of the 10th International symposium of Chinese CHI (Chinese CHI) / International Workshop on Human-Engaged computing (IWHEC)

作者： Liang, Xiaozhan Hu, Yong Li, Chuchen Beihang Univ Sch New Media Art & Design Beijing Peoples R China Beihang Univ Yunnan Innovat Inst Beijing Peoples R China Beihang Univ State Key Lab Virtual Real Technol & Syst Beijing Peoples R China

ISBN: (纸本)9781450398695

In recent years, especially after the outbreak of COVID-19, the applications of distributed collaboration in mixed reality (MR) environment has developed rapidly. Due to the cutting-edge technology and the specialized nature of the equipment, many studies have focused on the design of tutorials in MR, but few studies have taken the factors of emotional design into consideration. Since the perception of the state among users is based on their avatar firstly, which is helpful for users' perceptions about collaborative content in novice guidance. Based on Norman's principles of reflective design and James' theory about the experience self, we built a MR virtual sandtable collaboration system, which contains three guiding parts that are the full practice of theoretical research. Furthermore, We generated three progressive levels of emotional solutions in guidance design and conducted user experience research to explore the impact of three different degrees of emotional design, summarizing the applicable paradigm of novice guidance of emotional design in MR distributed collaborative scenarios. Through the evaluation and analysis of SUS model and task load, we found that novice guidance with a higher emotional design level, which combines static graphics, dynamic hand models and conversational voice would result in a better user experience.

关键词： Human-centered computing Interaction design Interaction design theory Concepts and paradigms

来源：评论

学校读者我要写书评

暂无评论

Time-space trade-offs in population protocols for the majority problem

引用

distributed computing 2021年第2期34卷 91-111页

作者： Berenbrink, Petra Elsaesser, Robert Friedetzky, Tom Kaaser, Dominik Kling, Peter Radzik, Tomasz Univ Hamburg Vogt Kolln Str 3 D-22527 Hamburg Germany Univ Salzburg Jakob Haringer Str 2 A-5020 Salzburg Austria Univ Durham Lower Mountjoy South Rd Durham DH1 3LE England Kings Coll London Strand CampusBush House London WC2B 4BG England

Population protocols are a model for distributed computing that is focused on simplicity and robustness. A system ofnidentical agents (finite state machines) performs a global task like electing a unique leader or determining the majority opinion when each agent has one of two opinions. Agents communicate in pairwise interactions with randomly assigned communication partners. Quality is measured in two ways: the number of interactions to complete the task and the number of states per agent. We present protocols for the majority problem that allow for a trade-off between these two measures. Compared to the only other trade-off result (Alistarh et al. in proceedings of the 2015 acm symposium on principles of distributed computing, Donostia-San Sebastian, 2015), we improve the number of interactions by almost a linear factor. Furthermore, our protocols can be made uniform (working correctly without any information on the population sizen), yielding the first uniform majority protocols that stabilize in a subquadratic number of interactions.

关键词： distributed computing Majority Population protocols Stochastic processes

来源：评论

学校读者我要写书评

暂无评论

Integrating Parallel and distributed computing in Early CS Courses 2022

Integrating Parallel and Distributed Computing in Early CS C...

引用

proceedings of the 53rd acm Technical symposium on Computer Science Education V. 2

作者： Sheikh Ghafoor Sushil Prasad Charles Weems Tennessee Technological University Cookeville TN USA University of Texas San Antonio San Antonio TX USA University of Massachusetts Amherst Amherst MA USA

ISBN: (纸本)9781450390712

Parallel and distributed computing (PDC) has become pervasive in all aspects of computing, and thus it is essential that students include parallelism and distribution in the computational thinking that they apply to problem solving, from the very beginning. Computer science education is still teaching to a 20th century model of algorithmic problem solving. Sequence, branch, and loop are taught in our early courses as the only organizing principles needed for algorithms, and we invest considerable time in showing how best to sequentially process large volumes of data. All computing devices that students use currently have multiple cores as well as GPU in many cases. Most of their favorite applications use multiple cores and numbers of distributed processors. Often concurrency offers simpler solutions than sequential approaches. acm and ABET have recommended including PDC in the undergraduate CS curriculum. However, we are still teaching them to solve problems using sequential thinking. In this workshop we overview the key PDC concepts and provide examples of how they may naturally be incorporated in early CS classes. We will introduce plugged and unplugged curriculum modules that have been successfully integrated in existing CS classes at multiple institutions. We will highlight the upcoming summer training that we are organizing, for which we have funding to support attendance.

关键词： early computing class pdc education hpc education

来源：评论

学校读者我要写书评

暂无评论

Efficient Shuffle Management with SCache for DAG computing Frameworks 18

Efficient Shuffle Management with SCache for DAG Computing F...

引用

23rd acm SIGPLAN symposium on principles and Practice of Parallel Programming (PPoPP)

作者： Fu, Zhouwang Song, Tao Qi, Zhengwei Guan, Haibing Shanghai Jiao Tong Univ Shanghai Peoples R China

ISBN: (纸本)9781450349826

In large-scale data-parallel analytics, shuffle, or the cross-network read and aggregation of partitioned data between tasks with data dependencies, usually brings in large overhead. To reduce shuffle overhead, we present SCache, an open source plug-in system that particularly focuses on shuffle optimization. By extracting and analyzing shuffle dependencies prior to the actual task execution, SCache can adopt heuristic pre-scheduling combining with shuffle size prediction to pre-fetch shuffle data and balance load on each node. Meanwhile, SCache takes full advantage of the system memory to accelerate the shuffle process. We have implemented SCache and customized Spark to use it as the external shuffle service and co-scheduler. The performance of SCache is evaluated with both simulations and testbed experiments on a 50-node Amazon EC2 cluster. Those evaluations have demonstrated that, by incorporating SCache, the shuffle overhead of Spark can be reduced by nearly 89%, and the overall completion time of TPC-DS queries improves 40% on average.

关键词： distributed DAG Frameworks Shuffle Optimization

来源：评论

学校读者我要写书评

暂无评论

proceedings - Adaptive Resource Management and Scheduling for Cloud computing: 3rd Internatioal Workshop, ARMS-CC 2016, Held in Conjunction with acm symposium on principles of distributed computing, PODC 2016

Proceedings - Adaptive Resource Management and Scheduling fo...

引用

3rd Internatioal Workshop on Adaptive Resource Management and Scheduling for Cloud computing, ARMS-CC 2016

ISBN: (纸本)9781450342278

The proceedings contain 10 papers. The topics discussed include: computation offloading from mobile devices: can edge devices perform better than the cloud?;performance of approximate causal consistency for partially-replicated systems;modeling the scalability of real-time online interactive applications on clouds;the impact on the performance of co-running virtual machines in a virtualized environment;a gossip-based dynamic virtual machine consolidation strategy for large-scale cloud data centers;data management of sensor signals for high bandwidth data streaming to the cloud;cloud elasticity: going beyond demand as user load;and cloud live streaming system based on auto-adaptive overlay for cyber physical infrastructure.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Massively Parallel Methods for Dynamic Programming 2017

Efficient Massively Parallel Methods for Dynamic Programming

引用

49th Annual acm-SIGACT symposium on Theory of computing (STOC)

作者： Im, Sungjin Moseley, Benjamin Sun, Xiaorui Univ Calif Elect Engn & Comp Sci 5200 N Lake Rd Merced CA 95344 USA Washington Univ Dept Comp Sci & Engn 1 Brookings Dr St Louis MO 63130 USA Univ Calif Berkeley Simons Inst Theory Comp Berkeley CA 94702 USA Simons Inst Theory Comp Berkeley CA USA

ISBN: (纸本)9781450345286

Modern science and engineering is driven by massively large data sets and its advance heavily relies on massively parallel computing platforms such as Spark, MapReduce, and Hadoop. Theoretical models have been proposed to understand the power and limitations of such platforms. Recent study of developed theoretical models has led to the discovery of new algorithms that are fast and efficient in both theory and practice, thereby beginning to unlock their underlying power. Given recent promising results, the area has turned its focus on discovering widely applicable algorithmic techniques for solving problems efficiently. In this paper we make progress towards this goal by giving principles for simulating a large class of sequential dynamic programs in the distributed setting. In particular, we identify two key properties, monotonicity and decomposibility, which allow us to derive efficient distributed algorithms for problems possessing the properties. We showcase our framework by considering several core dynamic programming applications, Longest Increasing Subsequence, Optimal Binary Search Tree, and Weighted Interval Selection. For these problems, we derive algorithms yielding solutions that are arbitrarily close to the optimum, using O(1) rounds and (O) over tilde (n/m) memory on each machine where n is the input size and m is the number of machines available.

关键词： Massively parallel computing Dynamic programming

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：