The customer churn, otherwise known as subscriber loss, inclusive among one of the key challenges of the telecommunication industry (TCI). Forecasting churn gives rise to telecoms step-saving retentional strategies, s...
详细信息
There are numerous difficulties concerning cyber security in the smart grid, and a majority of such challenges faced across devices connected to the grid are in the form of nefarious attacks with the potential aim to ...
详细信息
The increasing demand for computational power in big data and machine learning has driven the development of distributed training methodologies. Among these, peer-to-peer (P2P) networks provide advantages such as enha...
详细信息
ISBN:
(纸本)9798350343946
The increasing demand for computational power in big data and machine learning has driven the development of distributed training methodologies. Among these, peer-to-peer (P2P) networks provide advantages such as enhanced scalability and fault tolerance. However, they also encounter challenges related to resource consumption, costs, and communication overhead as the number of participating peers grows. In this paper, we introduce a novel architecture that combines serverless computing with P2P networks for distributed training and present a method for efficient parallel gradient computation under resource constraints. Our findings show a significant enhancement in gradient computation time, with up to a 97.34% improvement compared to conventional P2P distributed training methods. As for costs, our examination confirmed that the serverless architecture could incur higher expenses, reaching up to 5.4 times more than instance-based architectures. It is essential to consider that these higher costs are associated with marked improvements in computation time, particularly under resource-constrained scenarios. Despite the cost-time trade-off, the serverless approach still holds promise due to its pay-as-you-go model. Utilizing dynamic resource allocation, it enables faster training times and optimized resource utilization, making it a promising candidate for a wide range of machine learning applications.
In the past, data processing was completed on a computer or a server. This is because the resources available at that time were limited. However, with the progress of technology, it is possible to use multiple compute...
详细信息
distributed Generators (DGs) are becoming more integrated into the existing power system due to the growing popularity of renewable energy sources. However, in contrast to synchronous generators in a conventional powe...
详细信息
The technology is evolving fast. The data is growing faster. Apache Hadoop comes up as a capable platform to foster this rapidly progressing data. However, it is quite evident that the framework for Hadoop has been de...
详细信息
ISBN:
(数字)9798331521349
ISBN:
(纸本)9798331521356
The technology is evolving fast. The data is growing faster. Apache Hadoop comes up as a capable platform to foster this rapidly progressing data. However, it is quite evident that the framework for Hadoop has been designed to manage large files. As soon as it has to face a lot of small files, the performance sinks. This further leads to metadata overheads, processing overheads, network clogging and a list of other issues. This paper proposes a novel framework for merging these small files into large files known as pools. These pools can be updated dynamically whenever subsequent datasets are fed into the system according to the empty space available. Dynamic file pooling significantly improves the storage efficiency of Hadoop distributed File System (HDFS). The results are obtained over a smaller dataset which can be scaled for over a million of files in real scenario.
The main goal of the work was aimed to create a parallel application using a multithreaded execution model, which will allow the most complete and efficient use of all available computing resources. At the same time, ...
详细信息
ISBN:
(纸本)9783030941413;9783030941406
The main goal of the work was aimed to create a parallel application using a multithreaded execution model, which will allow the most complete and efficient use of all available computing resources. At the same time, themain attention was paid to the issues of maximizing the performance of the multithreaded computing part of the application and more efficient use of available hardware. During the development process, the effectiveness of various methods of software and algorithmic optimization was evaluated, taking into account the features of the functioning of a highly loaded multithreaded application, designed to run on systems with a large number of parallelcomputing threads. The problem of loading all available computing resources at the moment was solved, including the dynamic distribution of the involved CPU cores/threads and the computing accelerators, installed in the system.
In this paper, we develop an Attention based Generative Adversarial Networks (AGAN) to augment image data for the purpose of robust training for efficient processing of the hyper spectral imaging. The AGAN model enabl...
详细信息
ISBN:
(数字)9798331521349
ISBN:
(纸本)9798331521356
In this paper, we develop an Attention based Generative Adversarial Networks (AGAN) to augment image data for the purpose of robust training for efficient processing of the hyper spectral imaging. The AGAN model enables generation of images from the sample images that helps in training the classifier and in this study a fundamental classifier namely a convolutional neural network is used. A robust training is conducted to test the accuracy of detecting the instances effectively using the dataset. The simulation shows that the proposed AGAN-CNN attains improved accuracy after robust training than the existing methods.
This work proposes a voltage balance and flexible power sharing control strategy for modular input-parallel-output-parallel (IPOP) AC-DC converters with high frequency link (HF-link) isolation. Besides cell-to-cell po...
详细信息
Task allocation is an important problem that is encountered in different distributedcomputing environments such as grid and cloud computing. In this paper we take a multi-Agent learning approach for task allocation i...
详细信息
暂无评论