Knowledge and intent large data integration machine The simple beauty of Hadoop

"IT168 Information" December 3, 2012 news, the opening of the Beijing HBTC (Hadoop and large data technology assembly 2012, the original Hadoop in China) technology gathering, gathered a large number of scholars, business users and technology leaders. The Conference promotes the open source spirit angle, the union international and the domestic Hadoop and the Big Data application academic personage and the successful enterprise, examines the large data technology ecosystem's present situation and the development tendency through the technical application, revolves around the large processing, the information retrieval, the content mining, the natural language understanding, the data visualization, the Geographic information system and other related technical topics to explore, fully promote Hadoop and large data technology in the domestic application practice.

Mr. Chen Zhaoyu, Chief advisor to Etu, conducted a keynote address entitled "Best Practices for enterprise application of Hadoop" at the current session of the General Assembly. In his speech, Mr. Chen Zhaoyu not only introduced the latest features of the ETU Appliance 2.0, but also shared with the participants an understanding of the industry solutions and applications developed on ETU Appliance 2.0, Let more people who are interested in big data but have doubts about it, fully understand how big data really landed.

Etu is the leading brand of Asia's large data solutions, focusing on the development of large data solutions in the telecommunications, finance, and electronics industries, and on March 28, 2012, officially released the first large data-All-in-one product--etu appliance based on Hadoop in Asia. In this HBTC 2012 event, Etu was intent on sharing with the guests of the scene the latest version of the technology-superior Hadoop large Data Integration Machine ETU appliance. This version of the Hadoop ecosystem based on the open source, in addition to the computational efficiency of the user to build the system 3~12 times, but also implemented a large-scale cluster automatic deployment, not only the Name Node system comprehensive HA, KERBEROS/LDAP integrated multi-tenant security, High-efficiency data collection, centralized management of clusters and other features, is a real enterprise-class plug-and-play Hadoop integrated machine products.

"Because Etu understands the pain of the Hadoop application developer and system administrator, our unique design allows ETU Appliance 2.0 to do the Hadoop cluster deployment, management, HA, security is simply difficult to go beyond, computing and network transmission efficiency of high price." This is Asia's pride, and first published in China. Etu Cus, the head of the HBTC, said at the 2012 conference.

Mr. Jason Shih, an expert on large data Systems architecture from ETU, Sh Hongliang, also joined the guests in the wonderful sharing of "the evolution of the deployment of the Hadoop security architecture to higher services", and Mr. Sh Hongliang believes that the current open source framework using Hadoop The organization and management of large volumes of data has become a trend to address rapid data processing and improve analytical capabilities. However, enterprises have higher requirements for data and cluster system security, separating sensitive information from different levels or access rights and isolating shared storage or computer equipment. In previous versions of Hadoop 0.20, Unix-like file permissions had mentioned and provided simple authentication mechanisms within the cluster, but lacked access control for each job queue, control of commits, and so on. To use the new security features, the ETU Appliance 2 is set up as a Kerberos network authentication protocol, enabling it to have strong authentication and authorization, and to ensure strict access control data, resources and isolation between the tasks running. Mr. SH Hongliang shared the infrastructure deployment details of the Kerberos-based Hadoop security cluster environment and how to achieve high-level services. Implement the interoperability of existing activity domains with external LDAP domains, and effectively reduce the administrator overhead for integration and operations, through the ETU appliance rapid Deployment and the Cross-domain authentication mechanism built into the automation system.

In May 2012, Etu was officially released in Beijing with the Etu appliance knowledge and intent data integration machine based on Hadoop. The same year October, but also for e-commerce industry characteristics of the introduction of targeted solutions to know the intent of the precise recommendation system Etu recommender. Mark Etu know intention to ETU appliance the large data integration machine for data processing platform, supporting a variety of industrial applications, "one-more vertical" product towards the initial establishment.

Large data originates from the Internet, so most people who use large data are also rooted in the Internet. In order to improve the "talent bottleneck is the technical bottleneck" of the status quo, Etu to better help Chinese users realize the large data processing and analysis of Hadoop value, especially in Beijing set up a Hadoop development of the original team and technical support team, ETU know the intent of a large database to reduce Hadoop Adoption of the system thresholds, to help enterprises across the large data processing infrastructure barriers. It is the core meaning of large data that Etu know the intent of knowing the technology of Hadoop is the technical means to realize large data, let the enterprise discover the value behind the data, and concentrate on finding new business opportunities through data.