Volunteer computing (VC) is a type of networkcomputing which exploits idle computing resources provided by vast amount of users on the Internet. In our previous work, we have implemented a prototype system of paralle...
详细信息
ISBN:
(纸本)9798350386851;9798350386844
Volunteer computing (VC) is a type of networkcomputing which exploits idle computing resources provided by vast amount of users on the Internet. In our previous work, we have implemented a prototype system of parallel VC with the approach of server assisted communication and demonstrated the feasibility of parallelcomputing on a VC environment, not distributed computing as in the current VC. In this paper, we evaluate the effectiveness of parallel VC. The nature of nodes' volatility and unreliability in VC makes parallelcomputing not always effective. Through the simulations using open trace data of nodes' behavior, we reveal parallelism and redundancy of job execution which minimize execution time while maintaining high degree of reliability.
Datacenter is a big computing infrastructure that aims to reduce operating costs and energy consumption. Software-Defined networking (SDN) is a novel approach of networking that seeks to segregate control plane from d...
详细信息
ISBN:
(纸本)9798331515935;9783903176690
Datacenter is a big computing infrastructure that aims to reduce operating costs and energy consumption. Software-Defined networking (SDN) is a novel approach of networking that seeks to segregate control plane from data plane in order to reduce the complexity of network resource management. Optical Burst Switching (OBS) is an energy-efficient switching technology that offers flexibility to support datacenter traffic. OBS separates also control and data planes and can be the most compatible optical switching paradigm with SDN. This paper proposes an optical datacenter network architecture based on OBS and aligned with SDN specifications. The results obtained from simulation demonstrate the efficacy of this architecture, achieving zero burst loss, high throughput, and low burst delay.
Signed social networks, characterized by positive and negative attributes on their edges, offer a nuanced view of relationships between users. However, most existing datasets are outdated and fail to capture current s...
详细信息
ISBN:
(纸本)9798350391961;9798350391954
Signed social networks, characterized by positive and negative attributes on their edges, offer a nuanced view of relationships between users. However, most existing datasets are outdated and fail to capture current social dynamics. Additionally, as an important type, neutral links are not considered in most datasets. Hence, this paper introduces a novel signed social network dataset based on recent comments from YouTube videos in 2024. Unlike most traditional datasets, our dataset includes three types of relationships-positive, negative, and neutral-providing a more comprehensive representation of user interactions. Thus, our dataset can provide a more updated and diverse data for signed network research.
Extended reality offers unprecedented learning and training occasions, and unique challenges related not only to throughput and delay, but also to the characteristic spatial concentration of trainees. We have develope...
详细信息
ISBN:
(纸本)9798350390605;9783903176638
Extended reality offers unprecedented learning and training occasions, and unique challenges related not only to throughput and delay, but also to the characteristic spatial concentration of trainees. We have developed an algorithm for eXtended reality oriented Orchestration of Access Resources (X-OAR) grounding on next generation network technologies. X-OAR is designed to efficiently allocate edge computing facilities and cooperatively scheduled radio access network resources for extended reality applications. Building on the 3GPP guidelines on quality of experience in XR services, X-OAR meets the stringent XR delay requirements by leveraging edge and radio resources and employing cooperative scheduling within the radio access network. We introduce a graph model of the X-OAR optimization problem, and we present the X-OAR greedy algorithm, that reduces the orchestration complexity and the dependency on user subscription information. Experimental results show that X-OAR, with its cooperative scheduling technique, outperforms state-of-the-art competitors in terms of XR quality of experience. X-OAR paves the way for further studies extending the system orchestration to the application layer and the related resource charging policy.
Packet loss can seriously degrade the performance of TCP flows, especially that of small latency-sensitive flows. Managing packet loss efficiently is critical for maintaining good network performance. However, th...
详细信息
Private Edge computing (PEC) in IoT extends cloud services to the network edge to enhance applications. PEC faces challenges such as device heterogeneity, bandwidth constraints, and high latency. In this paper, we pro...
详细信息
ISBN:
(纸本)9798350390605;9783903176638
Private Edge computing (PEC) in IoT extends cloud services to the network edge to enhance applications. PEC faces challenges such as device heterogeneity, bandwidth constraints, and high latency. In this paper, we propose an end-edge collaborative clustered edge computing using stochastic drift federated optimization (SDFO). SDFO uses federated learning to train models locally in private isomorphic clusters, leveraging the power of edge devices to solve the device and data heterogeneity (Non-IID) problem and reduce reliance on central servers. SDFO optimizes the particle swarm algorithm through iterative optimization by transmitting only the relevant training parameters to reduce communication costs. Experimental results on CIFAR-10 and MNIST datasets validate the effectiveness of SDFO in handling non-IID data, reducing cost, and improving convergence and accuracy.
More and more portable intelligent devices are connected to the Internet in recent years. A way to effectively use the isolated cyber data without involving privacy and realize the cyber intrusion anomaly detection on...
详细信息
ISBN:
(纸本)9798350395716;9798350395709
More and more portable intelligent devices are connected to the Internet in recent years. A way to effectively use the isolated cyber data without involving privacy and realize the cyber intrusion anomaly detection on the portable intelligent devices with relatively limited hardware storage resources and computing power is worth exploring. In this paper, we propose a framework of federated anomaly detection, which enables the device effectively detect the anomaly by sharing the parameters of the federated model in a fully distributed fashion. We formulate the model training problem as a distributed robust optimization problem and subsequently devise an efficient algorithm for it. Experimental studies have also been carried out to reveal the superior performance of the proposed framework and underscore the significant benefits of federated anomaly detection.
The wireless sensor network is a hot and significant research area nowadays and it can be addressed in almost every sector and environment. A few major challenges are countered in the wireless sensor network such as e...
详细信息
ISBN:
(纸本)9798350372977;9798350372984
The wireless sensor network is a hot and significant research area nowadays and it can be addressed in almost every sector and environment. A few major challenges are countered in the wireless sensor network such as energy consumption, battery lifetime, Attacks, Data Transmission, etc. Generally wireless sensor network produces non-Euclidian sensing data and metadata structures and it is very complex to deal with the structure, especially in order to measure anomalies and disruption in a network. In this paper, we have introduced parallelcomputing to resolve the heterogeneity of the sensing data and metadata as well. parallelcomputing has been applied implicitly for extracting only paramount data from the large scale of data to detect routing layer attacks. The convolution neural network (CNN) has been considered as a machine-learning model and we have enhanced the kernel to optimize the performance of the conventional CNN model to detect network layer attacks in terms of wireless sensor networks. Our investigated method demonstrates better results in detecting anomalies and attacks than the existing methods or techniques.
In this research paper, we explore the essential role of High Performance computing (HPC) in the current technological era, highlighting its extensive use in various sectors, while also considering growing alarm over ...
详细信息
ISBN:
(纸本)9798350363074;9798350363081
In this research paper, we explore the essential role of High Performance computing (HPC) in the current technological era, highlighting its extensive use in various sectors, while also considering growing alarm over its environmental footprint. High performance computing systems are essential for managing large data sets and solving complex challenges. However, their significant contribution to escalating energy consumption and carbon emissions in the information and communication technology (ICT) sector cannot be ignored. Our study identifies the increasing energy demands and environmental challenges associated with HPC activities, including resource use, electrical waste and greenhouse gas emissions. It highlights the importance of understanding these environmental impacts in detail. It also contributes to the ongoing dialogue on sustainable computing, promoting a harmonious future where technological progress and environmental sustainability can coexist in unison.
In this paper, we present parallelo parallel Library (PPL), a novel Rust library for structured parallel programming. Rust's fearless concurrency concept makes it a promising language for parallel application deve...
详细信息
ISBN:
(纸本)9798350363074;9798350363081
In this paper, we present parallelo parallel Library (PPL), a novel Rust library for structured parallel programming. Rust's fearless concurrency concept makes it a promising language for parallel application development. We present the development progress of parallelo parallel Library (PPL) and report preliminary performance results, comparing it to existing popular Rust libraries. Our results show that PPL provides robust support for parallel programming through high-level abstractions, delivering performance that matches or exceeds current state-of-the-art in the Rust ecosystem.
暂无评论