检索结果-内蒙古大学图书馆

5th Workshop on Intelligent Data - From Data to Knowledge, DOING 2024, 3rd Workshop on Knowledge Graphs Analysis on a Large Scale, K-GALS 2024, 6th Workshop on Modern Approaches in Data Engineering and Information System Design, MADEISD 2024 and 3rd Workshop on Personalization and Recommender Systems, PERS 2024 held in conjunction with 28th European conference on Advances in databases and Information Systems, ADBIS 2024

作者： Barillaro, Luca Data Analytics Research Center Department of Medical and Surgical Sciences University "Magna Græcia" of Catanzaro Catanzaro88100 Italy

ISBN: (纸本)9783031704208

This paper aims to introduce my PhD research project, which is focused on scalable deep learning and its applications in the medical context. This project aims to design new DL algorithms or to adapt existing ones, to scalable architectures (e.g., parallel computers, GPUs), to improve the performance of typical ML tasks, such as classifications, and to experiment with them in the analysis of biomedical data, such as bioimages or molecular data. These applications are widely present in current literature thus representing a challenging opportunity. In addition to common DL and HPC approaches, the use of edge devices (i.e., Nvidia Jetson) is being explored, since this may be useful in a medical context using some key features of the edge computing paradigm (such as keeping data near the source) that are significant, i.e., for privacy reasons or legal compliance (i.e., European GPDR). My main investigated applications are on bioimages, such as Computed Tomography (CT) or functional Magnetic resonance imaging (fMRI). Research is being carried out to investigate current methodologies and how to improve them possibly. Moreover, a lot of experiments are carried out to demonstrate the impact of these approaches on traditional medical strategies. This work resulted in some experiments on the classification of medical images (CTs), medical signals (ECG), and gene expression data. My current work’s primary focus, boosted by a three-month collaboration with the University of Groningen, involves an extensive project on classifying fMRIs using Machine Learning techniques and concepts from graph theory tailored to exploit High-Performance Computing (HPC) infrastructures. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Gene expression

来源：评论

学校读者我要写书评

暂无评论

FPGA-based Real-time High-resolution Image Processing for rocket on-board processing

FPGA-based Real-time High-resolution Image Processing for ro...

引用

international conference on Computer Research and Development (ICCRD)

作者： Hui Zhou Xiaofeng Wang YuJia Xie Yongxiang Cao Yu Tang Yongchun Huo Huan Wen Xiongbo Zhao Chaoran Li Beijing Aerospace Automatic Control Institute National key Laboratory of Science and Technology on Aerospace Intelligent Control Beijing China Computer Science and Engineering Beihang University Beijing China National Superior College for Engineers Beihang University Beijing China

ISBN: (数字)9798331531881

ISBN: (纸本)9798331531898

The structure of the rocket-borne model is inherently complex, with processed images exhibiting high resolution and generating substantial amounts of data and calculations. Achieving robust real-time computing on an embedded platform poses significant challenges due to strictly limited resources, power consumption constraints, and size limitations. Our review of rocket-borne applications reveals considerable variability in the design resources of different devices, indicating a need for expanded design approaches. Upon evaluating existing methods, we identified two primary drawbacks. First, certain operators within the high-resolution target detection model are difficult to parallelize, resulting in significant inference delays that hinder the ability to meet task requirements. Although existing methods have been extended, there remains significant potential for performance enhancement in core scheduling for poor acceleration. This paper proposes an optimized architecture for the target detection algorithm accelerator designed for high-resolution images, along with a novel highly parallel data pre-processing and post-processing module implemented on FPGA to address these issues. Compared to the ARM implementation, this architecture demonstrates an improved performance of 24.64x. Furthermore, to ensure flexible application across various rocket launch scenarios, we introduce an optimization structure for convolution, pooling, and fusion operators and a multi-core expansion optimization method. This approach yields a 1.29x improvement in computing unit utilization compared to state-of-the-art multi-core scaling efforts. Finally, we assessed the accelerator architecture across multiple FPGA platforms, achieving a peak processing element utilization rate of 99.71% for a single core and layer. The overall computing efficiency, excluding the first layer, exceeded 90%. The peak computing power for the four cores reached 1638.4 GOPS, and the end-to-end computation time for

关键词： YOLO Rockets Power demand Convolution Multicore processing Reviews Scalability Accelerator architectures Real-time systems Field programmable gate arrays

来源：评论

学校读者我要写书评

暂无评论

PARBASE-90 international conference on databases, parallel Architecture and applications

PARBASE-90 International Conference on Databases, Parallel A...

引用

1990年

作者： Naphtali D Rishe Shamkant Bhalchandra Navathe Doron Tal

来源：评论

学校读者我要写书评

暂无评论

Graphical processing units and scientific applications

引用

international JOURNAL OF HIGH PERFORMANCE COMPUTING applications 2012年第3期26卷 189-191页

作者： Di Martino, Beniamino Mehofer, Eduard Quinlan, Dan Schordan, Markus Seconda Univ Napoli I-81031 Aversa CE Italy Univ Vienna A-1010 Vienna Austria Lawrence Livermore Natl Lab ROSE Compiler Project Livermore CA 94550 USA Univ Appl Sci Technikum Wien Vienna Austria

This special issue of the international Journal of High Performance Computing applications collects extended versions of the best three papers presented at the international Workshop on GPUs and Scientific applications (GPUScA 2010) held in Vienna in September 2010, in conjunction with PACT 2010 - the Annual international conference on parallel architectures and Compilation Techniques.

关键词： international(headers) journals parallel architectures High Performance Computing CONJUNCTION international conference Vienna Austria Collected papers

来源：评论

学校读者我要写书评

暂无评论

COMMONALITY IN architectures

COMMONALITY IN ARCHITECTURES

引用

2ND international SPECIALISTS SEMINAR ON THE DESIGN AND APPLICATION OF parallel DIGITAL PROCESSORS

作者： NICOLE, DA Univ of Southampton United Kingdom

ISBN: (纸本)0852965192

VLSI microprocessors have become central components in parallel computer architectures. parallel computers have reached a level of commercial viability for applications including databases and scientific modeling. This commercial imperative has also inspired staightforward designs that can make extensive use of existing components, both hardware and software. Multiprocessor architectures are converging. For the present, it is imperative to adopt common standards for message passing programming. For the future, it is expected that scalable virtual shared memory machines will dominate.

关键词： Computer Architecture

来源：评论

学校读者我要写书评

暂无评论

Throttling and speculating on parallel architectures

Throttling and speculating on parallel architectures

引用

PARBASE-90: international conference on databases, parallel architectures, and Their applications

作者： DeGroot, Doug Texas Instruments Computer Sci Cent

Throttling of parallelism is of importance to dynamic parallelism models in which the number and sizes of possible parallel code segments (tasks, processes, threads, etc.) are unknown at compile time, and which, if left uncontrolled, may lead to explosive parallelism with resulting slowdown rather than speedup. The primary goal of throttling is preventing a dynamically unfolding parallel execution from creating so many parallel code segments that either (1) the system runs out of resources with which to manage the parallel segments, (2) the system begins to thrash as a result of the increased process management and communication requirements, or (3) the system simply wastes CPU time creating extra code segments which end up being executed sequentially anyway. On the other hand, speculative parallelism involves the parallel execution of code segments which are not yet known to be required. It can be exploited in two basic forms: parallel search and parallel test. Speculative parallelism can be used when there are idle resources and the currently executing set of codes is not expected to yield any more parallel threads of execution.

关键词： Computer Systems, Digital

来源：评论

学校读者我要写书评

暂无评论

Predicting the behaviour of three-tiered applications:: dealing with distributed-object technology and databases

引用

PERFORMANCE EVALUATION 2000年第1-4期39卷 207-233页

作者： Savino-Vázquez, NN Anciano-Martin, JL Dumas, S Corbacho, JA Puigjaner, R Boudigue, D Gardarin, G Univ Illes Balears Dept Ciencies Matemat & Informat E-07071 Palma de Mallorca Spain Univ Simon Bolivar Dept Computac & Tecnol Informac Caracas 1080A Venezuela IFATEC F-78067 St Quenin En Yvelines France Univ Versailles Lab PRiSM F-78035 Versailles France World Trade Ctr TransTOOLS CA E-41092 Seville Spain

Today's applications are both object-oriented and based on a new type of three-tiered client-server architecture with clients, processing servers, and data servers as cornerstones. By recognising these trends, industry and researchers have been engaged in defining standards and technologies for communicating the components of Distributed Information Systems and for providing compatible mechanisms to access databases, but a key problem with these complex architectures is still their performance. This paper presents a tool for predicting the performance of systems based on CORBA and DCOM as distributed-object architectures, and OLE-DB and PL/SQL as data-access architectures. The tool is an extension of SMART, a workbench that exploits analytical and simulation performance models to predict the performance of database applications. (C)2000 Elsevier Science B.V. All rights reserved.

关键词： three-tiered architectures queuing-network models simulation models distributed-object architectures databases

来源：评论

学校读者我要写书评

暂无评论

A portable platform for parallel databases

A portable platform for parallel databases

引用

international conference and Exhibition on High-Performance Computing and Networking

作者： Stamatelopoulos, F Manis, G Papakonstantinou, G Digital Systems and Computer Laboratory National Technical University of Athens Greece Network Management and Optimal Design (NETMODE) Laboratory National Technical University of Athens Greece

ISBN: (纸本)3540593934

One of the most important topics in parallel application development is portability. No standard parallel computational model exists, whereas the multitude of different parallel architectures, programming paradigms and methodologies introduces significant difficulty in developing multi-system parallel applications. In this paper we present a platform-based architecture for developing portable parallel applications through the use of the platform approach. A platform is built on top of the native operating systems offering a common application programming interface (API). The API consists of a set of primitives through which portable applications can be developed (since only the platform - not the applications - requires re-programming for porting to different architectures). Furthermore, we define a set of key primitives specifically designed for implementing database applications. Our work towards this direction has been triggered by our recent implementation experience on parallel platforms and our interest in developing a parallel database system.

关键词： parallel architectures

来源：评论

学校读者我要写书评

暂无评论

parallel relational database systems: Why, how and beyond

Parallel relational database systems: Why, how and beyond

引用

7th international conference on Database and Expert Systems applications (DEXA 96)

作者： Hameurlain, A Morvan, F Lab. IRIT Université Paul Sabatier 31062 Toulouse 118 Route de Narbonne France

ISBN: (纸本)354061656X

This paper aims to describe synthetically integration and use of parallelism in relational databases on MIMD parallel architecture models. More precisely, after exposing the main goals of parallel relational databases, we highlight that It is essential to exploit recent parallel architectures to obtain high performance. parallelization of database programs requires the use of data placement approaches and data partitioning strategies which lead to extract levels, forms and types of parallelism. As for the inter-operation parallelization phase, the key problem of optimization, we describe one-phase and two-phase inter-operation parallelization strategies. This leads to unsolved problems which constitute a challenge for future parallel relational database systems.

关键词： parallel architectures

来源：评论

学校读者我要写书评

暂无评论

Beyond databases, architectures, and Structures: 10th international conference, BDAS 2014, Ustron, Poland, May 27-30, 2014. Proceedings 1

引用

丛书名： Communications in Computer and Information Science 424

2014年

作者： Stanislaw Kozielski, Dariusz Mrozek, Pawel Kasprowski, Bożena Małysiak-Mrozek, Daniel Kostrzewa (eds.)

ISBN: (数字)9783319069326

ISBN: (纸本)9783319069319;9783319069326

This book constitutes the refereed proceedings of the 10th IEEE international conference Beyond databases, architectures, and Structures, BDAS 2014, held in Ustron, Poland, in May 2014. This book consists of 56 carefully revised selected papers that are assigned to 11 thematic groups: query languages, transactions and query optimization; data warehousing and big data; ontologies and semantic web; computational intelligence and data mining; collective intelligence, scheduling, and parallel processing; bioinformatics and biological data analysis; image analysis and multimedia mining; security of database systems; spatial data analysis; applications of database systems; Web and XML in database systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：