The proceedings contain 49 papers presendted at a virtual meeting. The special focus in this conference is on parallel and distributedcomputing. The topics include: Data Management Model to Program Irregular Compute ...
ISBN:
(纸本)9783031061554
The proceedings contain 49 papers presendted at a virtual meeting. The special focus in this conference is on parallel and distributedcomputing. The topics include: Data Management Model to Program Irregular Compute Kernels on FPGA: Application to Heterogeneous distributed System;a distributed Game-Theoretic Approach to IaaS Cloud Brokering;continuous Self-adaptation of Control Policies in Automatic Cloud Management;Porting Sparse Linear Algebra to Intel GPUs;low-Overhead Reuse Distance Profiling Tool for Multicore;model-Based Loop Perforation;collaborative, distributed, Scalable and Low-Cost Platform Based on Microservices, Containers, Mobile Devices and Cloud Services to Solve Compute-Intensive Tasks;Scalable Hybrid parallel ILU Preconditioner to Solve Sparse Linear Systems;communication Overlapping Pipelined Conjugate Gradients for distributed Memory Systems and Heterogeneous Architectures;application-Based Fault Tolerance for Numerical Linear Algebra at Large Scale;parallelization and Auto-scheduling of Data Access Queries in ML Workloads;A Low Overhead Tasking Model for OpenMP;memory Efficient Deep Neural Network Training;Interferences Between Communications and Computations in distributed HPC Systems;enabling Support for Zero Copy Semantics in an Asynchronous Task-Based Programming Model;FleCSI 2.0: The Flexible Computational Science Infrastructure Project;An Experimental Study of SYCL Task Graph parallelism for Large-Scale Machine Learning Workloads;understanding the Effect of Task Granularity on Execution Time in Asynchronous Many-Task Runtime Systems;OpenMP Target Task: Tasking and Target Offloading on Heterogeneous Systems;towards Generating Realistic Trace for Simulating Functions-as-a-Service;SMART: A Tool for Trust and Reputation Management in Social Media;SPIRIT: A Microservice-Based Framework for Interactive Cloud Infrastructure Planning;consistency Analysis of distributed Ledgers in Fog-Enhanced Blockchains;parallelizing Automatic Model Management S
Marginal emissions rates-the sensitivity of carbon emissions to electricity demand-are important for evaluating the impact of emissions mitigation measures. Like locational marginal prices, locational marginal emissio...
详细信息
ISBN:
(纸本)9798350318562;9798350318555
Marginal emissions rates-the sensitivity of carbon emissions to electricity demand-are important for evaluating the impact of emissions mitigation measures. Like locational marginal prices, locational marginal emissions rates (LMEs) can vary geographically, even between nearby locations, and may be coupled across time periods because of, for example, storage and ramping constraints. This temporal coupling makes computing LMEs computationally expensive for large electricity networks with high storage and renewable penetrations. Recent work demonstrates that decentralized algorithms can mitigate this problem by decoupling timesteps during differentiation. Unfortunately, we show these potential speedups are negated by the sparse structure inherent in power systems problems. We address these limitations by introducing a parallel, reversemode decentralized differentiation scheme that never explicitly instantiates the solution map Jacobian. We show both theoretically and empirically that parallelization is necessary to achieve non-trivial speedups when computinggrid emissions sensitivities. Numerical results on a 500 node system indicate that our method can achieve greater than 10x speedups over centralized and serial decentralized approaches.
Generally, the single GPU computing method is utilized for the conventional radix sort algorithm based on GPU parallelcomputing. Nevertheless, as the data scale grows, the single GPU sorting algorithm is gradually de...
详细信息
ISBN:
(数字)9798350391954
ISBN:
(纸本)9798350391961;9798350391954
Generally, the single GPU computing method is utilized for the conventional radix sort algorithm based on GPU parallelcomputing. Nevertheless, as the data scale grows, the single GPU sorting algorithm is gradually demonstrating its performance bottlenecks. In the paper, an efficient radix sort algorithm based on multi-GPU parallelcomputing is proposed, which implements a strategy of using different bucket classifications on multiple GPUs to improve the sorting performance and efficiency of large-scale datasets. With the multi-GPU parallelcomputing, more buckets may be used for data classification in one traversal, effectively reducing data sorting times, lowering time complexity, and improving sorting speed and throughput. The experiment shows that the algorithm has significantly improved the operational efficiency, demonstrating good application prospects. Meanwhile, the algorithm herein also presents good scalability, which can adapt to the constantly growing data scale in the future.
Aimed at the distributed wind turbine integrating energy storage planning in distribution network, an optimal planning approach considering the different benefits of distribution company and distributed generation inv...
详细信息
In recent years, with the rapid development of the Internet of Things, the Internet, and social networks, the storage of data in the network is growing at an explosive rate and is becoming more and more closely relate...
详细信息
The proceedings contain 58 papers. The special focus in this conference is on parallel and distributedcomputing, Applications and Technologies. The topics include: Traffic Matrix Prediction Based on Differential...
ISBN:
(纸本)9783030967710
The proceedings contain 58 papers. The special focus in this conference is on parallel and distributedcomputing, Applications and Technologies. The topics include: Traffic Matrix Prediction Based on Differential Privacy and LSTM;patient-Chain: Patient-centered Healthcare System a Blockchain-based Technology in Dealing with Emergencies;federated Data Integration for Heterogeneous Partitions Based on Differential Privacy;multimodal Fusion Representation Learning Based on Differential Privacy;bayesian Optimization-Based Task Scheduling Algorithm on Heterogeneous System;MOFIT: An Efficient Access Control Scheme with Attribute Merging and Outsourcing Capability for Fog-Enhanced IoT;enhanced Discriminant Local Direction Pattern Learning for Robust Palmprint Identification;jointly Super Resolution and Degradation Learning on Unpaired Real-World Images;roman Amphitheater Classification Using Convolutional Neural Network and Data Augmentation;Multi-zone Residential HVAC Control with Satisfying Occupants’ Thermal Comfort Requirements and Saving Energy via Reinforcement Learning;Evaluating the Performance and Conformance of a SYCL Implementation for SX-Aurora TSUBASA;distributed Fair k-Center Clustering Problems with Outliers;MACSQ: Massively Accelerated DeepQ Learning on GPUs Using On-the-fly State Construction;optimizing Data Locality by Executor Allocation in Reduce Stage for Spark Framework;low Latency Execution Guarantee Under Uncertainty in Serverless Platforms;an Effective and Reliable Cross-Blockchain Data Migration Approach;A MVCC Approach to parallelizing Interoperability of Consortium Blockchain;realtime Physics Simulation of Large Virtual Space with Docker Containers;temperature Matrix-Based Data Placement Using Improved Hungarian Algorithm in Edge computing Environments;FastDCF: A Partial Index Based distributed and Scalable Near-Miss Code Clone Detection Approach for Very Large Code Repositories;matching Program Implementations and Heterogeneous computing System
distributed optimization algorithms have revolutionized the decision-making process in distribution network management. Unlike their centralized counterparts, the effectiveness of distributed algorithms is significant...
详细信息
ISBN:
(纸本)9798350318562;9798350318555
distributed optimization algorithms have revolutionized the decision-making process in distribution network management. Unlike their centralized counterparts, the effectiveness of distributed algorithms is significantly affected by the non-ideal states of communication networks used for data exchange. Hence, evaluating the resilience of distributed algorithms to communication imperfections is essential. Regarding this, this work examines the effectiveness of a distributed algorithm, named alternating direction method of multipliers (ADMM), in a three-phase unbalanced distribution system under the network delays of three popular cellular communication technologies. The performance assessment of the algorithm is done using the IEEE 123-bus test system by optimally scheduling the inverters connected to the network to achieve voltage deviation minimization. This analysis offers valuable understanding about the efficacy of the ADMM algorithm in terms of the solution quality, number of iterations, convergence rate and update frequency.
Affected by load balancing and synchronization and mutual exclusion problems among threads, domestic supercomputing platforms are prone to reduced parallel efficiency. In this study, MPI combined with OpenMP hybrid pa...
详细信息
With the rapid development of information technology, people have put forward higher requirements for the audio-visual experience and usage functions of conference spaces. The traditional, single-function conference r...
详细信息
ISBN:
(纸本)9798350391961;9798350391954
With the rapid development of information technology, people have put forward higher requirements for the audio-visual experience and usage functions of conference spaces. The traditional, single-function conference room configuration can no longer adapt to the diverse needs of modern work and interactive activities. How to achieve efficient management and control, seamless interconnection, and resource sharing of audio systems across spaces, while ensuring the acoustic characteristics and flexibility of each independent space, has become a core issue that needs to be urgently resolved in building a multi-hall, multi-functional conference room cluster. Based on the audio system project of the Academic Center of the Communication University of China, this paper proposes a solution for a conference room cluster audio system based on a distributed architecture. This solution not only overcomes the limitations of traditional systems in terms of scalability, collaborative work, and resource scheduling but also promotes lossless transmission, real-time processing, and adaptive configuration of audio signals. Thus, it ensures the consistency and high quality of the audio experience within the entire cluster, providing practical guidance for the design and optimization of future conference room audio systems.
With the increasing penetration of electric vehicles (EV) and distributed generations (DG) in distribution networks, operation and control of distribution networks (DN) is faced with many new challenges. Considering t...
详细信息
With the increasing penetration of electric vehicles (EV) and distributed generations (DG) in distribution networks, operation and control of distribution networks (DN) is faced with many new challenges. Considering the remarkable characteristics of distribution network layered by voltage level and the spatial and temporal distribution of EV charging load, a hierarchical and distributed optimization method for DN is proposed. Firstly, the spatial and temporal distribution prediction model of EV charging load is established, which is composed of three parts: the resident travel probability model, the vehicle mobility model and traffic network model. Secondly, these three parts runs jointly to simulate the charging demand generated by EV, and then the charging load is connected to the charging station in the low-voltage distribution station area. Thirdly, considering the operation characteristics of the new energy units, energy storage system (ESS), DG and EV charging station, a dynamic economic dispatching model is established. Fourthly, according to the hierarchical operation characteristics of the medium- and low-voltage distribution network (MLV), a distributed optimization algorithm of DN is proposed to realize the hierarchical and decoupled calculation, which transforms the traditional centralized serial computing mode into distributedparallelcomputing mode, and improves the computing efficiency of the system. At last, the effectiveness and superiority of the proposed strategy are verified by the simulation test. (c) 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under theCCBY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).
暂无评论