检索结果-内蒙古大学图书馆

2022 International Conference on Artificial Intelligence and Intelligent Information Processing, AIIIP 2022

作者： Gu, Liang Sun, Haichuan Zhao, Xixi Wang, Liguo Information and Communication Branch of State Grid Shanxi Electric Power Company China

ISBN: (纸本)9781510660175

In the era of big data, the network is always full of massive data. Cloud computing provides huge technical support for processing massive data. The cloud environment stores a large number of important information of individuals, enterprises and even countries. It has high commercial value, so it has become the primary target of many network attacks. Therefore, it is necessary to monitor the traffic in the cloud environment in real time and block abnormal traffic in time to ensure a safe and stable network environment for users. The existing intrusion detection systems can be divided into software systems and hardware systems, which are deployed in the backbone network in the network environment for real-time traffic detection. It is difficult to meet the traffic detection of multi branches and massive data in the cloud environment. This paper combines the deep neural network model with Hadoop framework, and proposes a distributed intrusion detection system model based on CNN-GRU. The deep neural network model is deployed in multiple nodes in the cloud environment, and the data is stored through HDFS and mapped and integrated by MapReduce method, so as to realize the intrusion detection of multi node parallel cloud environment. Finally, through the open source intrusion detection data set, the experimental results prove the effectiveness of the proposed method. © 2022 SPIE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

DARTS: distributed IoT Architecture for real-time, Resilient and AI-Compressed Workflows

DARTS: Distributed IoT Architecture for Real-Time, Resilient...

引用

2022 workshop on Advanced Tools, Programming Languages, and PLatforms for Implementing and Evaluating Algorithms for distributed systems, ApPLIED 2022

作者： Gupta, Ragini Chen, Bo Liu, Shengzhong Wang, Tianshi Sandha, Sandeep Singh Souza, Abel Nahrstedt, Klara Abdelzaher, Tarek Srivastava, Mani Shenoy, Prashant Smith, Jeffrey Wigness, Maggie Suri, Niranjan University of Illinois at Urbana-Champaign Urbana ChampaignIL United States University of California Los Angeles Los AngelesCA United States University of Massachusetts Amherst AmherstMA United States Devcom Army Research Laboratory AdelphiMD United States

ISBN: (纸本)9781450392808

IoT (Internet of Things) sensor devices are becoming ubiquitous in diverse smart environments, including smart homes, smart cities, smart laboratories, and others. To handle their IoT sensor data, distributed edge-cloud infrastructures are emerging to capture, distribute, and analyze them and deliver important services and utilities to different communities. However, there are several challenges for these IoT-edge-cloud infrastructures to provide efficient and effective services to users: (1) how to deliver real-time distributed services under diverse IoT devices, including cameras, meteorological and other sensors;(2) how to provide robustness and resilience of distributed services within the IoT-edge-cloud infrastructures to withstand failures or attacks;(3) how to handle AI workloads are in an efficient manner under constrained network conditions. To address these challenges, we present DARTS, which is composed of different IoT, edge, cloud services addressing application portability, real-time robust data transfer and AI-driven capabilities. We benchmark and evaluate these services to showcase the practical deployment of DARTS catering to application-specific constraints. © 2022 ACM.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

Understanding and Mitigating Hardware Interference Channels on Heterogeneous Multicore

Understanding and Mitigating Hardware Interference Channels ...

引用

real-time and Intelligent Edge Computing workshop (RAGE), IEEE

作者： Heechul Yun University of Kansas USA

ISBN: (数字)9798350363340

ISBN: (纸本)9798350363357

Contention for shared microarchitectural resources is a significant challenge for using modern heterogeneous multicore processors in safety-critical real-time applications. In this talk, I will first discuss major microarchitectural interference channels in modern multicore architecture. I will describe why microarchitectural designs aiming to achieve high memory-level parallelism (MLP), such as non-blocking cache and banked cache/DRAM organizations, could also become problematic interference channels from a real-time perspective. To illustrate their potential dangers, I will explain several effective “attack” strategies that have been shown to cause massive cross-core interference in real multicore platforms. Next, I will focus on the industrial challenge put forth by ARM in 2022, which is based on an augmented reality head-up display case-study from the automotive domain. I will discuss the impact of “aggressor” workloads on the performance of the case study's real-time applications on a real heterogeneous multicore platform (Jetson Nano). I will then present our software-based solution to achieve desired real-time performance without resorting to over-provisioning. Lastly, I will discuss the limitations of our approach and the need for better hardware support.

关键词： Microarchitecture Program processors Multicore processing Head-up displays Organizations parallel processing real-time systems

来源：评论

学校读者我要写书评

暂无评论

SSFuzz: State Sensitive Fuzzing for Network Protocol Implementations

SSFuzz: State Sensitive Fuzzing for Network Protocol Impleme...

引用

International Symposium on parallel and distributed Processing with Applications, ISPA

作者： Chengnuo Cai Lei Zhou Bo Yu School of Computer National University of Defense Technology China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

Protocol implementations are stateful and reactive systems, where the protocol process communicates with the client through a session. Complex state changes of implementations limit the efficiency of fuzzing. Current methods typically send a mutated message sequence to the protocol implementation, then terminate the session and evaluate the message sequence based on the feedback. We observe the following issues: (1) The state changes of the protocol implementation are random because of mutated messages, and message schedule methods that do not consider the real-time state of the target are blind; (2) The value of a message can vary significantly in different states, making seeds value evaluation method based on the entire message sequence insufficiently precise. (3) Existing fuzzing approaches actively terminate sessions after a test, resulting in resource *** solve these problem, we propose SSFuzz, a state sensitive fuzzing approach for protocol implementations. First, it monitors the state transitions and path feedback of the protocol implementation in real-time. Second, it dynamically schedules messages based on real-time state transitions during testing. According to the state transitions and path feedback after processing a message, it then conducts more precise evaluation of the messages. Last, we achieve testing multiple states within one session by reusing session. We believe that our approach better adapts to the characteristics of protocol implementations. We validated our approach on 9 widely used protocol implementations from ProFuzzBench. Compared to the state-of-the-art network protocol greybox fuzzing tool AFLnet, SSFuzz can increase discovered branch coverage on average 1.31% and discovers 25.28% more unique crashes within 24 hours.

关键词： Schedules distributed processing Protocols Fuzzing real-time systems Computer crashes Monitoring Testing

来源：评论

学校读者我要写书评

暂无评论

Testing of High-Lift System Functions with a distributed Avionics Test Bench 41

Testing of High-Lift System Functions with a Distributed Avi...

引用

41st IEEE/AIAA Digital Avionics systems Conference, DASC 2022

作者： Chrysalidis, Philipp Halle, Martin Thielecke, Frank Hamburg University of Technology Institute of Aircraft Systems Engineering Hamburg Germany

ISBN: (数字)9781665486071

ISBN: (纸本)9781665486071

The new emerging trends in aviation like more autonomous aircraft or green aviation lead to a higher development workload of the supporting avionic infrastructure. To support these technologies, new processes must be developed as to increase the efficiency with which market readiness is made available. Geographically distributed testing reduces the need for transportation of test benches during development and allows for easier adaption before, during and after tests. The communication can be conducted via real-time supporting systems, like the specifically designed quasi-real-time Internet connection hardware FastWAN developed by Exxpertsystems. Additionally, abstract simulation environments can be used as a further approach as to, e.g., confirm interface definitions. This allows for parallel development of the proper test setup while ensuring continuous validation. Therefore, faults can be isolated earlier, and overall development time is *** not replacing the need for proper certification testing, due to delays in the communication between the different systems, basic functionality and compatibility of subsystems can be analyzed. Consequently, development stakeholders are able to participate in the test process from their preferred development environment and therefore retain a high amount of flexibility. This increases the efficiency of late-stage tests, since the common test base is better *** research the advantages and challenges that this approach offers, a test bench for a high-lift system was set up and operated from three different locations distributed in Germany, each of them providing partial functionality. © 2022 IEEE.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

A Dynamic distributed Scheduler for DNN Inference on the Edge

A Dynamic Distributed Scheduler for DNN Inference on the Edg...

引用

parallel Computing and distributed systems (PCDS), International Symposium on

作者： Fei Hu Shivakant Mishra Department of Computer Science University of Colorado Boulder CO USA

ISBN: (数字)9798350349658

ISBN: (纸本)9798350349665

Edge computing plays a pivotal role in IoT applications that require rapid and secure data processing. How-ever, these applications are typically resource-demanding, and the resources available at the edge are often significantly less than those available in cloud environments. Thus, optimally utilizing these constrained resources is key to fulfilling various application requirements such as low latency, privacy, and cost-effectiveness. Given the dynamic and hybrid characteristics of IoT environments, static scheduling approaches frequently fail to adequately address these complex and varying demands. This paper presents the design, implementation, and evaluation of a dynamic distributed scheduler tailored for DNN inferences on edge computing platforms, based on sophisticated sub-model profiling techniques. This scheduler's core feature is its ability to continuously monitor the state of the IoT infrastructure, dynam-ically adapting the distribution of computing tasks according to real-time environmental and network conditions. Diverging from previous research that typically segments DNN models into two parts, this study proposes a more granular approach that splits models into multiple segments, thereby maximizing the utilization of diverse edge computing resources. The implementation of this scheduler, including integration with a GPU, demonstrates its effectiveness across various DNN models, highlighting its practical utility in leveraging edge computing capabilities.

关键词： Adaptation models Processor scheduling Computational modeling Graphics processing units Prototypes Dynamic scheduling real-time systems Internet of Things Resource management Edge computing

来源：评论

学校读者我要写书评

暂无评论

JORA: JAX Tensor-parallel LoRA Library for Retrieval Augmented Fine-Tuning 62

JORA: JAX Tensor-Parallel LoRA Library for Retrieval Augment...

引用

62nd Annual Meeting of the Association-for-Computational-Linguistics (ACL) / Student Research workshop (SRW)

作者： Tahir, Anique Cheng, Lu Liu, Huan Arizona State Univ 699 S Mill Ave Tempe AZ 85287 USA Univ Illinois 851 S Morgan St Chicago IL USA

ISBN: (纸本)9798891760967

The scaling of Large Language Models (LLMs) for retrieval-based tasks, particularly in Retrieval Augmented Generation (RAG), faces significant memory constraints, especially when fine-tuning extensive prompt sequences. Current open-source libraries support full-model inference and fine-tuning across multiple GPUs but fall short of accommodating the efficient parameter distribution required for retrieved context. Addressing this gap, we introduce a novel framework for PEFT-compatible fine-tuning of GPT models, leveraging distributed training. Our framework uniquely utilizes JAX's just-in-time (JIT) compilation and tensor-sharding for efficient resource management, thereby enabling accelerated fine-tuning with reduced memory requirements. This advancement significantly improves the scalability and feasibility of fine-tuning LLMs for complex RAG applications, even on systems with limited GPU resources. Our experiments show more than 12x improvement in runtime compared to Hugging Face/DeepSpeed implementation with four GPUs while consuming less than half the VRAM per GPU.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

A Scalable Platform for distributed Object Tracking Across a Many-Camera Network

引用

IEEE TRANSACTIONS ON parallel AND distributed systems 2021年第6期32卷 1479-1493页

作者： Khochare, Aakash Krishnan, Aravindhan Simmhan, Yogesh Indian Inst Sci Dept Computat & Data Sci Bangalore 560012 Karnataka India Indian Inst Sci Bangalore 560012 Karnataka India VMWare Bangalore 560076 Karnataka India

Advances in deep neural networks (DNN) and computer vision (CV) algorithms have made it feasible to extract meaningful insights from large-scale deployments of urban cameras. Tracking an object of interest across the camera network in near real-time is a canonical problem. However, current tracking platforms have two key limitations: 1) They are monolithic, proprietary and lack the ability to rapidly incorporate sophisticated tracking models, and 2) They are less responsive to dynamism across wide-area computing resources that include edge, fog, and cloud abstractions. We address these gaps using Anveshak, a runtime platform for composing and coordinating distributed tracking applications. It provides a domain-specific dataflow programming model to intuitively compose a tracking application, supporting contemporary CV advances like query fusion and re-identification, and enabling dynamic scoping of the camera network's search space to avoid wasted computation. We also offer tunable batching and data-dropping strategies for dataflow blocks deployed on distributed resources to respond to network and compute variability. These balance the tracking accuracy, its real-time performance, and the active camera-set size. We illustrate the concise expressiveness of the programming model for four tracking applications. Our detailed experiments for a network of 1000 camera-feeds on modest resources exhibit the tunable scalability, performance, and quality trade-offs enabled by our dynamic tracking, batching, and dropping strategies.

关键词： Cameras Streaming media Urban areas Tracking Cloud computing Target tracking Scalability Big data platform edge and fog computing video analytics distributed stream processing Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Reinforcement Learning with Partner Modeling for distributed Multi-agent Cooperation

引用

IEEE Transactions on parallel and distributed systems 2024年

作者： Liang, Zhixuan Cao, Jiannong Jiang, Shan Xu, Huafeng The Hong Kong Polytechnic University Department of Computing Hong Kong

Many real-world applications can be formulated as multi-agent cooperation problems, such as network packet routing and coordination of autonomous vehicles. The emergence of deep reinforcement learning (DRL) provides a promising approach for multi-agent cooperation through the interaction of the agents and environments. However, traditional DRL solutions suffer from the high dimensions of multiple agents with continuous action space during policy search. Besides, the dynamicity of agents' policies makes the training non-stationary. To tackle these issues, we propose HERO, a distributed hierarchical reinforcement learning approach that reduces training time and improves model performance by decomposing the overall policy into multiple layers of sub-policies in a hierarchical manner. On the one hand, the cooperation of multiple agents can be learned efficiently in high-layer discrete action space. On the other hand, the low-layer individual control can be reduced to single-agent reinforcement learning. Furthermore, we introduce a partner modeling network to effectively capture and model other agents' policies during learning. Finally, We develop a real-world multi-vehicle testbed and evaluate HERO in a series of cooperative lane change scenarios. Both the simulation and real-world experiments show that our approach can achieve a lower average collision rate of 0.09 and a faster convergence speed than other baselines. © 1990-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A knowledge-driven multiobjective algorithm for distributed hybrid flowshop with group and carryover setup in glass manufacturing systems

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2023年第1期181卷

作者： Geng, Ya-Dian Li, Jun-Qing Yunnan Normal Univ Dept Math Kunming 650500 Peoples R China HengXing Univ Sch Informat Engn Qingdao 266199 Peoples R China

The distributed hybrid flow shop problem (DHFSP) exists widely in real workshop production processes. Its process optimization has important academic significance and application value, and has become one of the research hotspots in the field of production scheduling. In this study, a multiobjective DHFS problem with carryover setup time and group constraints (DHFGSP) is considered with the background of glass processing, and the objectives are the completion time and time of use price. First, a mathematical programming model is established according to the processing flow of glass production. Second, an improved nondominated sorting genetic algorithm-II (INSGAII) is adopted to optimize the problem. Third, the evolutionary strategies based on problem knowledge are developed, and a local search strategy based on critical factories is designed. Finally, the proposed algorithm is compared with several of the most advanced multiobjective algorithms, which shows that INSGA-II has better performance for solving DHFGSP.

关键词： distributed hybrid flowshop group scheduling problem Multiobjective Carryover setup time Speed-up procedure Glass manufacturing system

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：