This research focuses on the study of network security protection techniques for the central station side of distributed power dispatch and control systems. It covers various aspects such as vulnerability discovery, i...
详细信息
distributed photovoltaic systems are one of the key technologies for achieving China39;s carbon peaking and carbon neutrality goals, with their continuous development and technological progress being crucial. This s...
详细信息
Super pixel algorithm SLIC uses K-means mean clustering method to effectively generate super pixels. Compared with other super pixel algorithms, it is more efficient and improves the segmentation performance. In order...
详细信息
The proceedings contain 15 papers. The special focus in this conference is on Southwest Data Science. The topics include: How Does Normalization Impact Clustering?;quantitative Stock Market Modeling Using Multivariate...
ISBN:
(纸本)9783031678707
The proceedings contain 15 papers. The special focus in this conference is on Southwest Data Science. The topics include: How Does Normalization Impact Clustering?;quantitative Stock Market Modeling Using Multivariate Geometric Random Walk;disease Similarity and Disease Clustering;composition Analysis and Identification of Ancient Glass Products;data Entropy-Based Imbalanced Learning;analysis of the Changes in Microbial Community During the Fermentation of Feng-Flavour Baijiu;enhanciGraph: Visualizing Enhancer-Gene Interactions;event-Triggered Control for Synchronization of Chaotic Delayed Lur’e systems with Stochastic Cyber-Attacks;in-Game Win Prediction Models for Cricket;finite-Time Outer Average Synchronization Between Two Coupled Heterogeneous Complex Dynamical Networks and Its Application in Secure Communication;Classification of In-Situ Solar Wind Data Measured by Solar Orbiter/SWA-PAS and HIS Using Machine Learning;node Classification with Multi-hop Graph Convolutional Network;pyDaskShift: Automatically Convert Loop-Based Sequential Programs to distributedparallel Programs.
In recent years, Approximate Nearest Neighbor Search (ANNS) has played a pivotal role in modern search and recommendation systems, especially in emerging LLM applications like Retrieval-Augmented Generation. There is ...
详细信息
ISBN:
(纸本)9798400704369
In recent years, Approximate Nearest Neighbor Search (ANNS) has played a pivotal role in modern search and recommendation systems, especially in emerging LLM applications like Retrieval-Augmented Generation. There is a growing exploration into harnessing the parallel computing capabilities of GPUs to meet the substantial demands of ANNS. However, existing systems primarily focus on offline scenarios, overlooking the distinct requirements of online applications that necessitate real-time insertion of new vectors. This limitation renders such systems inefficient for real-world scenarios. Moreover, previous architectures struggled to effectively support real-time insertion due to their reliance on serial execution streams. In this paper, we introduce a novel Real-Time Adaptive Multi-Stream GPU ANNS System (RTAMS-GANNS). Our architecture achieves its objectives through three key advancements: 1) We initially examined the real-time insertion mechanisms in existing GPU ANNS systems and discovered their reliance on repetitive copying and memory allocation, which significantly hinders real-time effectiveness on GPUs. As a solution, we introduce a dynamic vector insertion algorithm based on memory blocks, which includes in-place rearrangement. 2) To enable real-time vector insertion in parallel, we introduce a multi-stream parallel execution mode, which differs from existing systems that operate serially within a single stream. Our system utilizes a dynamic resource pool, allowing multiple streams to execute concurrently without additional execution blocking. 3) Through extensive experiments and comparisons, our approach effectively handles varying QPS levels across different datasets, reducing latency by up to 40%-80%. The proposed system has also been deployed in real-world industrial search and recommendation systems, serving hundreds of millions of users daily, and has achieved significant results.
This article combines the Gaussian mixture model with the extended propagation single particle filtering technique (ESPT) and proposes a Gaussian mixture distributed particle filtering algorithm that uses ESPT for sta...
详细信息
distributed self-adaptive system is a multi-component collaborative system that automatically adjusts its behavior and structure through adaptive mechanisms to maintain system performance and stability in dynamic envi...
详细信息
This study introduces two innovative approaches to augment intrusion detection capabilities. The first method employs Principal Component Analysis (PCA) for dimensionality reduction, streamlining the dataset39;s com...
详细信息
With the increasing integration of renewable energy into the power grid, the demand for efficient communication systems for managing and controlling distributed power sources has become urgent. Here, we introduce the ...
详细信息
This paper explores using artificial intelligence in applications to detect tomato leaf diseases. A key issue is how automated systems and artificial intelligence can be used to diagnose and predict disease control, d...
详细信息
暂无评论