High-performance serverless computing has garnered significant attention. Researchers have developed numerous optimization strategies for serverless frameworks to fully leverage the benefits of serverless computing. H...
详细信息
With the expansion of the use scope of electric power equipment, the prevention and control of electrical fire is becoming more and more important. Traditional electrofire detection methods have some problems of trigg...
详细信息
Transitive closure computation is a fundamental operation in graph theory with applications in various domains. However, the increasing size and complexity of real-world graphs make traditional algorithms inefficient,...
详细信息
Motivated by a wide range of applications from parallelcomputing to distributed learning, we study distributed online load balancing among multiple workers. We aim to minimize the pointwise maximum over the workers&#...
详细信息
ISBN:
(纸本)9798350339864
Motivated by a wide range of applications from parallelcomputing to distributed learning, we study distributed online load balancing among multiple workers. We aim to minimize the pointwise maximum over the workers' local cost functions. We propose a novel algorithm termed distributed Online Load Balancing with rIsk-averse assistancE (DOLBIE), which jointly considers the worker heterogeneity and system dynamics. The workload is distributed to workers in an online manner, where the underloaded workers learn to provide an appropriate amount of assistance to the most overloaded worker for the next online round without making themselves overwhelmed. In DOLBIE, all workers participate in updating the workload simultaneously, and no computationally intensive gradient or projection calculation is required. DOLBIE can be implemented in both the master-worker and fully-distributed architectures. We analyze the worst-case performance of DOLBIE by deriving an upper bound on its dynamic regret. We further demonstrate the application of DOLBIE to online batch-size tuning in distributed machine learning. Our experimental results show that, in comparison with state-of-the-art alternatives, DOLBIE can substantially speed up the training process and reduce the workers' idle time.
Sparse matrix reordering is an important step in Cholesky decomposition. By reordering the rows and columns of the matrix, the time of computation and storage cost can be greatly reduced. With the proposal of various ...
详细信息
ISBN:
(纸本)9798350359329;9798350359312
Sparse matrix reordering is an important step in Cholesky decomposition. By reordering the rows and columns of the matrix, the time of computation and storage cost can be greatly reduced. With the proposal of various reordering algorithms, the selection of suitable reordering methods for various matrices has become an important research topic. In this paper, we propose a method to predict the optimal reordering method by visualizing sparse matrices in chunks in a parallel manner and feeding them into a deep convolutional neural network. The results show that the theoretical performance can reach 95% of the optimal performance, the prediction accuracy of the method can reach up to 85%, the parallel framework achieves an average speedup ratio of 11.35 times over the serial framework, and the performance is greatly improved compared with the traversal selection method on large sparse matrices.
The purpose is to ensure the big data acquisition of parallel battery back state, and the safe and effective operation of the energy management system. Edge devices are combined with cloud computing to achieve a big d...
详细信息
The purpose is to ensure the big data acquisition of parallel battery back state, and the safe and effective operation of the energy management system. Edge devices are combined with cloud computing to achieve a big data acquisition and processing model based on the edge computing, which makes the speed of the big data acquisition of parallel battery back state faster, and avoids data fitting. The algorithm is optimised based on the combination of the Lyapunov method and the distributed method of ADMM (Alternating Direction Multipliers Method). The optimised edge computing improves the performance of the energy management system of parallel battery back state. The experimental results show that the two methods can effectively avoid the fitting phenomenon of data acquisition, and the distributed method can simplify the complexity of data processing and make the energy management system consume minimum energy. The big data acquisition speed of parallel battery back state based on the improved edge computing is faster, and the battery energy management is more effective, which has enormous significance for enlarging the application of parallel battery.
Escalating global power(energy) demands and the need to avail it in a reliable, efficient manner has led to the modernization of legacy and current power system grids into Smart grid (SGs) equivalents his article prop...
详细信息
In recent years, with the progress of distributed photovoltaic technology and the reduction of costs, the installed scale of distributed photovoltaic has grown rapidly. Due to the high average photovoltaic permeabilit...
详细信息
Decentralized Multi-agent Learning (DML) enables collaborative model training while preserving data privacy. However, inherent heterogeneity in agents' resources (computation, communication, and task size) may lea...
详细信息
ISBN:
(纸本)9798350386066;9798350386059
Decentralized Multi-agent Learning (DML) enables collaborative model training while preserving data privacy. However, inherent heterogeneity in agents' resources (computation, communication, and task size) may lead to substantial variations in training time. This heterogeneity creates a bottleneck, lengthening the overall training time due to straggler effects and potentially wasting spare resources of faster agents. To minimize training time in heterogeneous environments, we present a Communication-Efficient Training Workload Balancing for Decentralized Multi-Agent Learning (ComDML), which balances the workload among agents through a decentralized approach. Leveraging local-loss split training, ComDML enables parallel updates, where slower agents offload part of their workload to faster agents. To minimize the overall training time, ComDML optimizes the workload balancing by jointly considering the communication and computation capacities of agents, which hinges upon integer programming. A dynamic decentralized pairing scheduler is developed to efficiently pair agents and determine optimal offloading amounts. We prove that in ComDML, both slower and faster agents' models converge, for convex and non-convex functions. Furthermore, extensive experimental results on popular datasets (CIFAR-10, CIFAR-100, and CINIC-10) and their non-I.I.D. variants, with large models such as ResNet-56 and ResNet-110, demonstrate that ComDML can significantly reduce the overall training time while maintaining model accuracy, compared to state-of-the-art methods. ComDML demonstrates robustness in heterogeneous environments, and privacy measures can be seamlessly integrated for enhanced data protection.
This study comprehensively sorts out the main tower foundation types used in overhead transmission line projects with voltage levels of 35kV and above, and then deduces the carbon emission calculation model and theore...
详细信息
暂无评论