Cloud computing has become a major and most preferred technology for data storage across the world. This technology tremendously improved service delivery to clients as data and services are brought closer to the clie...
详细信息
ISBN:
(纸本)9781665417501
Cloud computing has become a major and most preferred technology for data storage across the world. This technology tremendously improved service delivery to clients as data and services are brought closer to the client. Bringing data and services closer to the client requires the implementation of a distribution architecture, that is, an architecture that spans across a wider geographical area. However, such architecture comes with its own challenges. It requires proper management of data and services across multiple data centres to maintain consistency, availability of data and services, and systems.that are tolerant to failures.. In this study we focus on evaluating currently used methodologies to ensure data consistency performance levels for cloud storage and determine whether the proposed solutions are effective and feasible. From the surveyed data consistency models, we noted that it’s difficult to find one model that satisfies the three main challenges associated with data consistency across multiple geo-replicated data centres.
We present TaskTorrent, a lightweight distributed task-based runtime in C++.TaskTorrent uses a parametrized task graph to express the task DAG, and one-sided active messages to trigger remote tasks asynchronously. As ...
详细信息
Local search algorithms are widely applied in solving large-scale distributed Constraint Optimization Problems(DCOPs). In local search algorithms, each agent makes its decision based on values of its neighbours. Thus,...
详细信息
The proceedings contain 169 papers. The topics discussed include: impact of the geometric field of view on drivers39; speed perception and lateral position in driving simulators;analysis and processing of environmen...
The proceedings contain 169 papers. The topics discussed include: impact of the geometric field of view on drivers' speed perception and lateral position in driving simulators;analysis and processing of environmental monitoring system;on using physical based intrusion detection in SCADA systems.modern approach to design a distributed and scalable platform architecture for smart cities complex events data collection;fuzzy similarities for road environment-type detection by a connected vehicle from traffic sign probabilistic data;activity recognition in the city using embedded systems.and anonymous sensors;forecasting risk of diseases in Kazakhstan with using mapping technique based on 9 years statistics;understanding the relation between distance and train station choice behavior of cyclists in the western region of the Netherlands;door-to-door transit accessibility using Pareto optimal range queries;using neural nets to predict transportation mode choice: an Amsterdam case study;and a model of second-degree virtual cut nodes applied to complex networks in ecological landscape.
This article proposes a maximum power point tracking (MPPT) algorithm based on perturb observe (PO) method and sliding mode control (SMC). For efficient control of the PV system, the accurate and efficient model of th...
详细信息
MLPerf benchmarks, which measure training and inference performance of ML hardware and software, have published three sets of ML training results so far. In all sets of results, ResNet50v1.5 was used as a standard ben...
详细信息
ISBN:
(纸本)9781450383035
MLPerf benchmarks, which measure training and inference performance of ML hardware and software, have published three sets of ML training results so far. In all sets of results, ResNet50v1.5 was used as a standard benchmark to showcase the latest developments on image recognition tasks. The latest MLPerf training round (v0.7) featured Intel's submission with TensorFlow. In this paper, we describe the recent optimization work that enabled this submission. In particular, we enabled BFloatl6 data type in ResNet50v1.5 model as well as in Intel-optimized TensorFlow to exploit full potential of 3rd generation Intel Xeon scalable processors that have built-in BFloat16 support. We also describe the performance optimizations as well as the state-of-the-art accuracy/convergence results of ResNet50v1.5 model, achieved with large-scale distributed training (with upto 256 MPI workers) with Horovod. These results lay great foundation to support future MLPerf training submissions with large scale Intel Xeon clusters.
The proceedings contain 74 papers. The special focus in this conference is on Advancement of Science and Technology. The topics include: Earliest-Arrival Route: A Global Optimized Communication for Networked Control S...
ISBN:
(纸本)9783030806170
The proceedings contain 74 papers. The special focus in this conference is on Advancement of Science and Technology. The topics include: Earliest-Arrival Route: A Global Optimized Communication for Networked Control systems.Impulsive Noise Mitigation Using Hybrid Nonlinear Preprocessor and Turbo Code in OFDM-PLC System;comparative Analysis of Kalman Filtering and Machine Learning Based Cardiovascular Signal Processing Algorithms;modeling and Analysis of Three-Phase Inverter for Induction Motor Drive for Three-Wheel Electric Vehicle Application;performance Analysis of Multicarrier Modulation Techniques for Next Generation Networks;synchronous Generator Excitation Loss Detection Based on Reactive Power Flow Limit;estimation of Synchrophasor Parameters in the Presence of 3rd & 5th Harmonics and White Gaussian Noise;Performance Analysis of Vertical Sectorization in Sub-6-GHz Frequency Bands for 4G Mobile Network Under Realistic Deployment Scenario;power Loss Reduction and Voltage Profile Improvement of Radial Distribution System Through Simultaneous Network Reconfiguration and distributed Generation Integration;A Comprehensive Review and Evaluation of Classical MPPT Techniques for a Photovoltaic System;antiviral Activity of Aloe pirottae A. Berger Root Extracts Against Influenza A and B Viruses, Picornaviruses and Dengue Virus: An Endemic Plant Species of Ethiopia;dynamic Performance Evaluation of Type-3 Compensator for Switch-Mode Dc-Dc Converter;inset-Feed Rectangular Microstrip Patch Antenna Array Performance Enhancement for 5G Mobile Applications;blockchain Based Green Coffee Supply Chain Management to Improve Traceability and Transparency (Case Study on Sidama Coffee);wireless Local Area Network Intrusion Detection System Using Deep Belief Networks;Direct Adaptive Fuzzy PI Strategy for a Smooth MPPT of Variable Speed Wind Turbines.
The proceedings contain 74 papers. The special focus in this conference is on Advancement of Science and Technology. The topics include: Earliest-Arrival Route: A Global Optimized Communication for Networked Control S...
ISBN:
(纸本)9783030806200
The proceedings contain 74 papers. The special focus in this conference is on Advancement of Science and Technology. The topics include: Earliest-Arrival Route: A Global Optimized Communication for Networked Control systems.Impulsive Noise Mitigation Using Hybrid Nonlinear Preprocessor and Turbo Code in OFDM-PLC System;comparative Analysis of Kalman Filtering and Machine Learning Based Cardiovascular Signal Processing Algorithms;modeling and Analysis of Three-Phase Inverter for Induction Motor Drive for Three-Wheel Electric Vehicle Application;performance Analysis of Multicarrier Modulation Techniques for Next Generation Networks;synchronous Generator Excitation Loss Detection Based on Reactive Power Flow Limit;estimation of Synchrophasor Parameters in the Presence of 3rd & 5th Harmonics and White Gaussian Noise;Performance Analysis of Vertical Sectorization in Sub-6-GHz Frequency Bands for 4G Mobile Network Under Realistic Deployment Scenario;power Loss Reduction and Voltage Profile Improvement of Radial Distribution System Through Simultaneous Network Reconfiguration and distributed Generation Integration;A Comprehensive Review and Evaluation of Classical MPPT Techniques for a Photovoltaic System;antiviral Activity of Aloe pirottae A. Berger Root Extracts Against Influenza A and B Viruses, Picornaviruses and Dengue Virus: An Endemic Plant Species of Ethiopia;dynamic Performance Evaluation of Type-3 Compensator for Switch-Mode Dc-Dc Converter;inset-Feed Rectangular Microstrip Patch Antenna Array Performance Enhancement for 5G Mobile Applications;blockchain Based Green Coffee Supply Chain Management to Improve Traceability and Transparency (Case Study on Sidama Coffee);wireless Local Area Network Intrusion Detection System Using Deep Belief Networks;Direct Adaptive Fuzzy PI Strategy for a Smooth MPPT of Variable Speed Wind Turbines.
Depth as well as intensity of a pixel plays a significant role in labeling objects in 3D environments. This paper presents a novel approach of labeling objects from multi-view video sequences by incorporating rich dep...
详细信息
ISBN:
(纸本)9789813292918;9789813292901
Depth as well as intensity of a pixel plays a significant role in labeling objects in 3D environments. This paper presents a novel approach of labeling objects from multi-view video sequences by incorporating rich depth information. The depth map of a scene is estimated from focus-cues using the Gaussian-Hermite moments (GHMs) of local neighboring pixels. It is expected that the depth map obtained from GHMs provides robust features as compared to that provided by other popular depth maps such as those obtained from Kinect and defocus cue. We use the rich depth and intensity values of a pixel to score every point of a video frame for generating labeled probability maps in a 3D environment. These maps are then used to create a 3D scene wherein available objects are labeled distinctively. Experimental results reveal that our proposed approach yields excellent performance of object labeling for different multi-view scenes taken from RGB-D object dataset, in particular showing significant improvements in precision-recall characteristics and F1-score.
The proceedings contain 11 papers. Topics discussed include: combining antibiotic with fog computing: antibiotic 2.0;towards context-aware and dynamic management of stream processing pipelines for fog computing;using ...
ISBN:
(纸本)9781728123653
The proceedings contain 11 papers. Topics discussed include: combining antibiotic with fog computing: antibiotic 2.0;towards context-aware and dynamic management of stream processing pipelines for fog computing;using virtual events for edge-based data stream reduction in distributed publish/subscribe systems.optimal placement of stream processing operators in the fog;development of a smart metering microservice based on fast Fourier transform (FFT) for edge/Internet of things environments;edge-to-edge resource discovery using metadata replication;enabling fog computing using self-organizing compute nodes;robust resource allocation using edge computing for vehicle to infrastructure (V2I) networks;ORCH: distributed orchestration framework using mobile edge devices;and machine learning based timeliness-guaranteed and energy-efficient task assignment in edge computingsystems.
暂无评论