Online Trends

Research on cloud computing platform based on parallel Gaussian elimination Pan Xiaohui in order to solve the problem that the Gaussian elimination algorithm of the serial partial selected principal element cannot make full use of the multi-core processor, a Gaussian elimination algorithm of the partial selected principal element of parallel multithreading is proposed and implemented, and the whole algorithm is analyzed and optimized, The storage layout of the data and the pattern of the algorithm are matched, which greatly improves the performance of the program. By comparing and analyzing the experimental results on the local Linux server and the various platforms of Amazon EC2 Cloud, the Gaussian elimination algorithm of the partial selected principal element is greatly affected by the cache, so it is in CP ...

Data distribution and storage design of ship information System Wang Kanmei Hu Xiaohui Zhou Wenya Gao with the increasing requirements of data interaction between various systems of ships, the explosive increase of historical data, the efficient distribution of data and the increasing importance of mass data storage. On the basis of analyzing the specification of DDS (data distribution service), the Ship Information data distribution system is built, and a joint storage scheme based on RDB and NoSQL (not only SQL) is proposed. To realize the data of ship information system ...

Operation management practice of large scale cloud data center China Telecom Cloud 2014-5-23 China Telecom Co., Ltd. Cloud Computing branch operations general manager Wu Wei published the theme of "Large-scale cloud data center operation Management Practice" speech. What should be the evolution of real cloud data centers in the coming years? Wu Wei that it should be built on the development of a data center that hosts the Internet. "Planning the entire Network cloud data center layout, the formation of the bearer network to form an integrated service standards and intensive operation mode, through the infrastructure to the application and the provision of data cloud services, upgrade services ...

A collaborative mechanism for computing and data in large data technology Wang Peng Huang Liu Fengan Handsome Data system is also known as a data-oriented high-performance computing system, similar to the traditional high-performance computing systems, and its computing and data storage is usually based on the cluster implementation of distributed systems. Based on the coordination mechanism of computation and data, this paper compares the high performance computing and data-oriented high performance computation, and points out that the cooperative mechanism of computing and data determines the basic structure and performance of large data system. The integration of Distributed file system and computation through assistance mechanism is the realization of automatic parallelization of large data system.

Spatial analysis in the age of large data Zhangxiaoxiang in recent years, large data are increasingly used to describe the coming era of information explosion, but also caused a high degree of attention by government departments. Firstly, the characteristics and latest development of large data and large space data are reviewed. Secondly, the core issues of three geo-information science, such as spatial analysis, spatial modeling and spatial optimization, are reviewed and reviewed from the perspective of GIS basic theory. On this basis, it extends the space philosophy and space thinking behind the space theory method. Finally, it points out the opportunity and challenge of the development of geo-information science in large data age. &nbs ...

Large data and productivity the Development Research Center of the State Council Lee Paoxi Big Data has opened a new era, it is changing our thinking and life and the way to understand the world, as a source of innovative production and services. When we study the future productivity development of our country, we must consider the influence of large data. Some experts say " Big data and cloud computing are two sides of a problem, one is the problem, the other is the solution to the problem. Therefore, to understand the big data must understand cloud computing. It can be said that cloud computing is a big data age of the system engineering. This paper studies the concept and characteristics of large data, and analyzes the large data cloud ...

Cloud workflow scheduling algorithm oriented to price dynamics Zhengmin Jian Yiao Yan a cloud workflow resource scheduling algorithm based on dynamic programming is proposed to solve the problem of cloud Workflow task scheduling cost optimization in the environment of Resource price change. The algorithm divides the tasks in the cloud workflow into multiple serial task groups and then uses the dynamic programming algorithm to schedule each task to calculate the scheduling scheme with the lowest cost in the task group. By comparing the experiment with the state Graph search algorithm, the scheduling results of four kinds of workflow models are calculated when the price of the dispatched resource is variable function.

Research and realization of large data management oriented to the field of Astronomy Harbin Institute of Technology Yao Hua This article will optimize the database performance of Oracle's real application cluster, testing to see if the existing Oracle real application cluster meets the needs of the project. In addition, this article will compare the performance of query statements in Oracle's real application cluster and other database management systems in the same situation. This article will also be tested on the scalability of Oracle's real application cluster to see if Oracle's real application cluster can support adding, expanding clusters to ...

The construction of Semantic spatial resource index and its query optimization algorithm based on object-oriented network Ma Wu Bin Liumingshing Dengsu Huanghongbin face the massive space resources in the Internet of things, introduce the method of semantic classification in the construction of Spatial resource Index, make the traditional IR tree have semantic feature, support the user's semantic query to the resource, A top-k semantic query optimization algorithm is proposed, and the validity of index structure and the correctness of query result are obtained through the use of actual data and experimental analysis, and the performance of the algorithm is evaluated from 3 indexes, such as index construction time, query time and query accuracy rate. Object-oriented Internet language ...

Format for manufacturing industry The cloud service platform architecture of document management Shantao Royan Liang The purpose of this paper is to analyze the functional structure of the manufacturing-oriented layout industrial document management system and how to combine with cloud computing to build an on-demand cloud platform for service. Methods according to the mature manufacturing-oriented layout industrial document management system functional requirements, from which can be combined with cloud computing functional requirements to make IT service, and then based on these services using sophisticated server virtualization technology to build a realistic demand for the cloud platform. Results after the document management system function is serviced, users can ...

Content-Aware trusted cloud one-stop industry solution case in the Second China Cloud computing conference large Data Entertainment industry application forum, the blue Flood ChinaCache chief technology officer and President Zhang Ken published "Content perception trusted Cloud One-stop industry solution" as the keynote speech,   This paper introduces the advantages of the blue flood content perception trusted Cloud one-stop service. Content-Aware trusted cloud one-stop industry solutions

The research and application of Enterprise cloud Computing in Angang mining industry Huan Handeju Bihonggang Xu Xin Yin Qingfang Angang Group Mining company self-developed "Mining Cloud" system, innovative memory extraction technology, so that the system reliability has been significantly improved and reliable protection; innovative shelling technology, compatible with all other virtual machines and operating systems, The first application of Xen technology, the implementation of cloud host, cloud storage, cloud services, cloud terminal applications. Through independent innovation, integrated innovation, for enterprises to build a smart "mining cloud" system platform. ...

Design and implementation of converged cloud computing platform Jiang Rei Xu Peng in the rapid development of information technology today, more and more enterprises have and began to build their own private cloud to manage computing resources, there are many emerging enterprises in the purchase of public cloud computing power to achieve rapid application of the agency. The cloud computing platform, especially the IaaS and PAAs Cloud platform, has made great progress both at home and abroad, but there are also some problems such as the inability of the scheduling ability to effectively integrate the resource information of IaaS and PAAs cloud platforms. This paper presents a fusion of cloud computing flat ...

The reason for the "cloud" at present, Anhui University has built Anhui province and even the entire east and central China region's first real education private cloud (IaaS and desktop virtualization), and Xiamen Institute of Technology deployed Xiamen first real meaning of the private cloud (IaaS). Talking about the original intention of implementing cloud computing, Anhui University Network Center director Shan told reporters: "As early as 2004, we have basically achieved the hardware concentration, the campus of most departments of the Department of server storage and other hardware centrally hosted in the school's data center, sharing the computer room environment and network resources." But that's just ...

What is a private cloud according to Gartner, companies are investing more in private cloud services than public cloud services in the next few years, but in the end, most will be transitioning from private cloud to public cloud. At the same time, the analysis agency said many companies will experience a private cloud in the future, a mixture of public and non-cloud systems, and the process may last for years. 　　This development route means that in the next few years, the primary responsibility of the enterprise IT department will be to find the most appropriate IT services for the enterprise. However, put aside the forecast does not talk about, at present ...

Research on the design of Cloud computing virtual exhibition platform in Shenyang International Convention and Exhibition Center Jilin Sui Airlines This article will elaborate on the platform's operating mode and commercial characteristics and the necessity of virtual exhibition. Including the research significance of virtual exhibition, domestic and foreign status quo and so on. Detailed talk about the cloud computing virtual exhibition using a variety of technologies, including virtual reality technology, VRML Technology, 3D Studio Max technology, JavaScript technology, augmented reality technology, mobile internet technology. Based on the integrated system ...

The iterative MapReduce framework for evolutionary algorithm Jin Weijian Wang Branch MapReduce modular programming greatly reduces the implementation difficulty of the distributed algorithm, but also limits its application scope. This paper introduces the basic structure of mapreduce and the defects of its iterative algorithm, and proposes an iterative MapReduce computing framework which is suitable for evolutionary algorithm based on the research of MapReduce's computational framework on the basis of the inefficiency of mapreduce evolutionary algorithm. Describes the iterative m ...

Design and implementation of outlier mining algorithm in data-intensive computing environment Chenyali Changrombo Li Yahong Zhang Shusen Liu Xi based on mapreduce model, this paper proposes a grid technology and a outlier mining algorithm LOF based on Mr_lof method. The map phase uses the grid for data reduction to send the representative point information to the master node, and the reduced phase uses the density based outlier mining algorithm to filter out the dense regions with the help of grid value E. The algorithm only needs to compute sparse area objects ...

Research on two-layer dynamically deployable virtual machine scheduling Strategy Ecust Jiang This paper proposes a new virtual machine scheduling framework to allocate the resources of cloud computing data center to users in a reasonable and efficient way. This scheduling strategy has made innovations in the following three areas. 1. Introduction of two-tier queuing mechanism. The first level queue is the license queue, which controls the user's access rights, and the second ranks are classified as processing queues, and the final processing priority is determined according to the type of user, the queue time and the proportion of the type of user. Using Fs-aloha algorithm to detect and deal with collisions, to ensure that ...

Analysis and implementation of cloud computing monitoring model for network public opinion Wu aiming at the shortcomings of traditional network public opinion monitoring, choosing a more mature cloud computing architecture, adopting a unified collection of LAN logs, and storing and analyzing the massive data using cloud computing technology, this paper proposes a kind of network public opinion cloud computing monitoring mode and its implementation method The experimental data of the equipment configuration and analysis efficiency are also given. Analysis and implementation of cloud computing monitoring model of network public opinion