检索结果-内蒙古大学图书馆

OneOS: Distributed Operating System for the Edge-to-Cloud Continuum

ieee TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2025年第6期36卷 1175-1192页

作者： Jung, Kumseok Gascon-Samson, Julien Gopalakrishnan, Sathish Pattabiraman, Karthik Univ British Columbia Dept Elect & Comp Engn Vancouver BC V6T 1Z4 Canada ETS Montreal Montreal PQ H3C 1K3 Canada

Application developers often need to employ a combination of software such as communication middleware and cloud-based services to deal with the challenges of heterogeneity and network dynamism in the edge-to-cloud continuum. Consequently, developers write extra glue code peripheral to the application's core business logic, to provide interoperability between interacting software frameworks. Each software framework comes with its own framework-specific API, and as technology evolves, the developer must keep up with the changing APIs by updating the glue code in their application. Thus, framework-specific APIs hinder interoperability and cause technology fragmentation. We propose a design of a middleware-based distributed operating system (OS) called OneOS to realize a computing paradigm that alleviates such interoperability challenges. OneOS provides a single system image of the distributed computing platform, and transparently provides interoperability between software components through the standard POSIX API. Using OneOS's domain-specific language, users can compose complex distributed applications from legacy POSIX programs. OneOS tolerates failures by adopting a distributed checkpoint-restore algorithm. We evaluate the performance of OneOS against an open-source IoT Platform, ThingsJS, using an IoT stream processing benchmark suite, and a video processing application. OneOS executes the programs about 3x faster than ThingsJS, reduces the code size by about 22%, and recovers the state of failed applications within 1 s upon detecting their failure.

关键词： Codes Cloud computing Logic Pipelines Costs Streaming media Standards programming Domain specific languages Interoperability N-tier architectures edge computing cloud computing development frameworks and environments domain specific languages

来源：评论

学校读者我要写书评

暂无评论

Constructing Temporal networks of OSS programming Language Ecosystems 30

Constructing Temporal Networks of OSS Programming Language E...

引用

30th ieee International conference on Software Analysis, Evolution and Reengineering (SANER)

作者： Agroskin, Alexander Lyulina, Elena Titov, Sergey Kovalenko, Vladimir Weizmann Inst Sci Rehovot Israel JetBrains Res Belgrade Serbia JetBrains Res Limassol Cyprus JetBrains Res Amsterdam Netherlands

ISBN: (纸本)9781665452786

One of the primary factors that encourage developers to contribute to open source software (OSS) projects is the collaborative nature of OSS development. However, the collaborative structure of these communities largely remains unclear, partly due to the enormous scale of data to be gathered, processed, and analyzed. In this work, we utilize the World Of Code dataset, which contains commit activity data for millions of OSS projects, to build collaboration networks for ten popular programming language ecosystems, containing in total over 290M commits across over 18M projects. We build a collaboration graph representation for each language ecosystem, having authors and projects as nodes, which enables various forms of social network analysis on the scale of language ecosystems. Moreover, we capture the information on the ecosystems' evolution by slicing each network into 30 historical snapshots. Additionally, we calculate multiple collaboration metrics that characterise the ecosystems' states. We make the resulting dataset publicly available% including the constructed graphs and the pipeline enabling the analysis of more ecosystems.

关键词： Ecosystems

来源：评论

学校读者我要写书评

暂无评论

Latency-Optimal network Microservice Architecture Deployment in SDN

Latency-Optimal Network Microservice Architecture Deployment...

引用

ieee conference on Global Communications (ieee GLOBECOM) - Intelligent Communications for Shared Prosperity

作者： Luis Herrera, Juan Scotece, Domenico Galan-Jimenez, Jaime Berrocal, Javier Di Modica, Giuseppe Foschini, Luca Univ Extremadura Dept Comp Syst & Telemat Engn Badajoz Spain Univ Bologna Dipartamento Informat Scienza Ingn Bologna Italy

ISBN: (纸本)9798350310900

The Software-Defined networking (SDN) paradigm enables network administrators to manage the behavior of the network thanks to a centralized control plane. By programming network-level applications, it is possible to determine how traffic flows must be handled programmatically. In the same manner that computing applications have evolved from monolithic to microservice-based architectures, network-level applications are expected to evolve into microservice-based SDN controllers, implementing each application as a replicable and individually deployable component of the SDN controller. In such a scenario, the Quality of Service (QoS) experienced by traffic flows depends on how these microservices are placed and deployed through the network topology. In this work, we provide a system to optimize the QoS of the traffic in microservice-based SDN networks by optimally placing and replicating the network-level microservices. Experimental results show the effectiveness of the proposed solution over a real network topology with varying traffic loads.

关键词： Software-Defined networking Service Oriented Computing Micro-services Quality of Service

来源：评论

学校读者我要写书评

暂无评论

network Slicing Design Policy for Computation Offloading using Deep Reinforcement Learning

Network Slicing Design Policy for Computation Offloading usi...

引用

2024 ieee conference on Standards for Communications and networking, CSCN 2024

作者： Wang, Ranyin Friderikos, Vasilis King's College London Department of Engineering Strand LondonWC2R 2LS United Kingdom

ISBN: (纸本)9798331507428

In the era of beyond 5G and 6G, mobile devices with limited resources are challenged by the rising computational demands of emerging intelligent applications. Task offloading offers a viable solution by transferring computational tasks from end devices to remote servers, thereby effectively reducing the processing burden on these devices. network slicing provides flexible and dynamic resource allocation, making it an ideal mechanism for facilitating computation offloading. This paper presents a network slicing framework in the scope of openRAN architecture to address computation offloading employing different Deep Reinforcement Learning (DRL) algorithms. We propose an intra-slice design policy for optimizing communication and computation resource allocations within slices, aiming to minimize offloading delays. Specifically, the Actor-Critic (AC), Deep Q-network (DQN), and Deep Deterministic Policy Gradient (DDPG) algorithms are leveraged to adaptively address the slicing design problem and generate intra-slice policies efficiently. Compared to baseline approaches, our proposed algorithms demonstrate rapid convergence and superior delay performance. © 2024 ieee.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

Integration, testing and performance analysis of ETSI-MEC with Different OAI 5G architectures 5

Integration, testing and performance analysis of ETSI-MEC wi...

引用

5th ieee India Council International Subsections conference, INDISCON 2024

作者： Sanadhya, Shashvat Gandotra, Pimmy Lall, Brejesh Iit Delhi Department of EE/BSTTM New Delhi India JIIT/IITD Department of ECE/BSTTM Noida/Delhi New Delhi India Iit Delhi Department of Ee New Delhi India

ISBN: (纸本)9798350376753

For the realization of flexible, customized, diverse 5G and beyond 5G (B5G) services, network slicing (NS) has emerged as a competent technology. Further, the multi-access edge computing (MEC) technology plays a key role in efficient management of the slices in the next generation networks, supporting compute close to the users. These gain further popularity with the rising diversity in the applications, seeking end-to-end service guarantees. For provisioning isolated end-to-end network slices with guaranteed services to different applications, this paper discusses network slicing test bed using the open Air Interface (OAI). An integration of the MEC and core in 5G are proposed in the OAI test bed set-up. The paper consolidates different network slicing set-ups, tested using the OAI. The objective for the same is to evaluate the effectiveness of OAI. The stringent latency requirements of the upcoming networks can be effectively met by a union of network slicing technology with MEC, as shall be analysed. © 2024 ieee.

关键词： Integration testing

来源：评论

学校读者我要写书评

暂无评论

Review of the SDN Architecture with Various API Controllers

Review of the SDN Architecture with Various API Controllers

引用

2023 ieee International conference on Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering, RMKMATE 2023

作者： Elumalai, J. Subhashini, R. Faculty of CSE Sathyabama Institute of Science and Technology India

ISBN: (纸本)9798350305708

SDN revolutionizes internetwork addressing, aiming to transcend the constraints of conventional configurations, facilitating adaptable network typologies. This trailblazing concept has gained remarkable traction in contemporary digital internetwork design. At its core, SDN merges control and data planes, yielding refined resource administration, elevated security, and comprehensive control. A spectrum of controllers, including floodlight, NOT, RYU, POX, open daylight (ODL), and open network operating systems (ONOS), has been engineered, collectively contributing to the architecture's constituents. This manuscript deeply probes into SDN network scalability, fault tolerance, and interfaces, concurrently conducting an extensive examination of pivotal facets within SDN architecture: API, controller infrastructures, modularity, extensibility, programming language, community scalability, telemetry, resilience, and fault tolerance. © 2023 ieee.

关键词： Floodlight OK open network Operating System (ONOS) openDaylight (ODL) RYU

来源：评论

学校读者我要写书评

暂无评论

Connectivity Schemas in NeuroEvolution: What Neural architectures does GEPNN evolve?

Connectivity Schemas in NeuroEvolution: What Neural Architec...

引用

2023 ieee Symposium Series on Computational Intelligence, SSCI 2023

作者： Jonathan, Mwaura Ryan, Heminway Khoury College of Computer Sciences Roux Institute at Northeastern University PortlandME United States Khoury College of Computer Sciences Northeastern University BostonMA United States

ISBN: (纸本)9781665430654

In recent years, there has been a rise in popularity of using evolutionary algorithms (EA's) in conjunction with artificial neural networks (ANN's). This approach is commonly known as NeuroEvolution (NE). NeuroEvolutionary approaches typically optimize just the weights of an ANN or optimize the architecture, learning rates, thresholds, and weights together. Algorithms capable of the latter are known as Topological and Weight Evolving ANN (TWEANN). One such TWEANN is Gene Expression programming for Neural networks (GEPNN). This paper presents an empirical investigation of the network topologies that arise when GEPNN is used and whether evolved architectures have any relation to state of the art architectures. Results show that GEPNN naturally discovers powerful structural motifs such as shortcut connections and also creates sparse networks. Both these schemas have been shown to be advantageous in deep learning techniques. As an additional contribution from this work, we provide an open source library for developing GEPNN solutions in Python. © 2023 ieee.

关键词： architectures Evolutionary computation intelligent systems neural networks NeuroEvolution

来源：评论

学校读者我要写书评

暂无评论

Novel Deep Learning network architectures for Time Series Forecasting

Novel Deep Learning Network Architectures for Time Series Fo...

引用

2023 ieee International conference on Paradigm Shift in Information Technologies with Innovative Applications in Global Scenario, ICPSITIAGS 2023

作者： Katariya, Jitendra Kumar Yadav, Dhananjay Gayathri, C. Kolekar, Vikas K. Daivagna, Umesh M. Pushparajesh, V. Vivekananda Global University Department of Computer Science & Application Jaipur India Maharishi School of Engineering and Technology Maharishi University of Information Technology UP India Prince Shri Venkateshwara Padmavathy Engineering College Department of Science and Humanities Chennai India Vishwakarma Institute of Information Technology Department of Computer Engineering Pune India Atlas SkillTech University Department of UGDX Maharastra Mumbai India Faculty of Engineering and Technology Department of Electrical and Electronics Engineering Karnataka Bangalore India

ISBN: (纸本)9798350329773

This paper presents novel deep-learning network architectures for time series forecasting. First, a singular deep gaining knowledge of network architecture is proposed and tested for the usage of the Google tendencies dataset as a benchmark. The proposed structure combines 1-dimensional convolution neural networks, recurrent neural networks, and long-term cells to seize and study lengthy-term styles in the time series information. The structure is compared to current deep getting-to-know architectures and is shown to outperform them in phrases of accuracy. Moreover, the proposed structure is applied to a secondary time collection dataset, the COVID-19 confirmed infection dataset, to attain compelling effects. Eventually, an open-source implementation of the proposed architecture is made available for use in addition to research. © 2023 ieee.

关键词： network architecture

来源：评论

学校读者我要写书评

暂无评论

Toward Highly Reliable Programmable Data Planes: Verification of P4 Code Generation 9

Toward Highly Reliable Programmable Data Planes: Verificatio...

引用

9th ieee International conference on network Softwarization (ieee NetSoft) - Boosting Future networks through Advanced Softwarization

作者： Gyoergyi, Csaba Laki, Sandor Schmid, Stefan Univ Vienna Vienna Austria Eotvos Lorand Univ Budapest Hungary TU Berlin Berlin Germany Fraunhofer SIT Darmstadt Germany

ISBN: (纸本)9798350399806

Data plane programming gained much attention in the past years, having a fast-growing community both in academia and industry. Many tools have emerged to simplify and/or help the development of reliable data plane programs, including fuzzing, formal verification, and different code generators. However, even the tools themselves must be verified to meet the most stringent dependability requirements. In this paper, we investigate various tools and methods to verify code generators leveraging P4 through the example of P4RROT (an open source code generator focusing on the application layer). We show that our approach is efficient and can indeed successfully find bugs. We identify two bugs and propose reusable ideas, such as the use of ghost code.

关键词： code generation in-network computing data plane verification P4

来源：评论

学校读者我要写书评

暂无评论

PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization

PracticalDG: Perturbation Distillation on Vision-Language Mo...

引用

ieee/CVF conference on Computer Vision and Pattern Recognition (CVPR)

作者： Chen, Zining Wang, Weiqiu Zhao, Zhicheng Su, Fei Men, Aidong Meng, Hongying Beijing Univ Posts & Telecommun Sch Artificial Intelligence Beijing Peoples R China Beijing Key Lab Network Syst & Network Culture Beijing Peoples R China Minist Culture & Tourism Key Lab Interact Technol & Experience Syst Beijing Peoples R China Brunel Univ Uxbridge Uxbridge Middx England

ISBN: (纸本)9798350353006

Domain Generalization (DG) aims to resolve distribution shifts between source and target domains, and current DG methods are default to the setting that data from source and target domains share identical categories. Nevertheless, there exists unseen classes from target domains in practical scenarios. To address this issue, open Set Domain Generalization (OSDG) has emerged and several methods have been exclusively proposed. However, most ex-isting methods adopt complex architectures with slight improvement compared with DG methods. Recently, vision-language models (VLMs) have been introduced in DG following the fine-tuning paradigm, but consume huge training overhead with large vision models. Therefore, in this paper, we innovate to transfer knowledge from VLMs to lightweight vision models and improve the robustness by introducing Perturbation Distillation (PD) from three perspectives, including Score, Class and Instance (SCI), named SCI- PD. Moreover, previous methods are oriented by the benchmarks with identical and fixed splits, ignoring the divergence between source domains. These methods are revealed to suffer from sharp performance decay with our proposed new benchmark Hybrid Domain Generalization (HDG) and a novel metric H-2-CV, which construct various splits to comprehensively assess the robustness of algorithms. Extensive experiments demonstrate that our method outperforms state-of-the-art algorithms on multiple datasets, especially improving the robustness when con-fronting data scarcity.

关键词： Visual languages

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：