Large data processing only Hadoop is not enough

Hadoop has largely been synonymous with big data in the enterprise market since Oracle, IBM and Microsoft in 2011 unanimously announced their support for Hadoop. Today, the situation may have changed.

NoSQL increasingly important

Because of the high profile of Hadoop, few people notice that in the same year that they announced their support for Hadoop, the three relational database vendors also announced the support of the NoSQL database, respectively.

As open source software, the birth and development of NoSQL (not just SQL) is also to meet the requirements of Web 2.0, especially the social network for the "three high" database, that is, the high concurrency of the database read and write, the massive data storage and access, high scalability of the database. These are just the short boards of relational data.

However, to put the NoSQL on the Internet, applied to the enterprise-level market a bit so that users can not completely put down their hearts. After all, the Internet rarely runs the key business applications, and which internet giant has not interrupted services, but downtime for internet companies, more of the gray-scale face, more embarrassing. But for those enterprise-class users who put their Gbagbo on it, their key business applications are not allowed to slip.

Oracle NoSQL database is divided into free and paid two editions while providing business support. As an extensible, highly available key-value database, Oracle NoSQL Database provides scalable key/value storage through integration with Oracle Event 處理 to efficiently capture real-time large data. "Sishing, vice president of Oracle Company and technical general manager of Greater China region, said

Sishing said that, based on the development of recent years, in the real enterprise-level application level, the paid commercial version is more recognized by customers. The visa company recently deployed NoSQL to detect the theft of swipe cards. "When many companies are developing to the Hadoop level, or when large data technologies bring some value, the industry needs a vendor to fully support all of the above architectures, which at the same time demonstrates Oracle's industry leadership in the big data dimension." With the coexistence and integration of Hadoop, NoSQL, and SQL, large data technologies will have a higher degree of maturity. ”

Overall solution large data

"Structured data and unstructured data coexist and contain each other, and the integration of the two processes constitutes a complete and large data processing." "Sishing that," after the acquisition of Sun, Oracle has the integration of hardware and software capabilities, and then form an integrated system to provide customers with large data processing capacity of the solution. ”

According to the introduction, NoSQL in unstructured processing, through the integration with Oracle Event 處理 (OEP) to achieve efficient capture of real-time large data, and then through the MapReduce method for processing and querying. Oep's application of real-time filtering, association and processing events helps to eliminate, integrate, correlate, and filter data, and to avoid clutter in the data warehouse, thus enabling real-time analysis of a large number of data streams.

After Hadoop emerged, Oracle built a large data processing platform through the Hadoop Enterprise Edition and Oracle NoSQL database.

Oracle then merged the Database Engine (DBE), storage, server and fiber to form a database cloud platform for Oracle Exadata to process structured data. The integration of Hadoop with Oracle database and Exadata is optimized through Oracle large data connectors, and the Hadoop system is seamlessly connected to Oracle database systems to enable unstructured and structured data stitching to form a complete large data solution.

In data analysis, Oracle exalytics business Intelligence cloud Servers are developed using industry standard hardware, business intelligence software, and memory database technology, the first integrated system specifically designed to provide high-performance analysis, modeling, Discovery, and planning. It loads and reads data from Oracle Exadata over a high-bandwidth InfiniBand bus.

In the decision Support phase, the information is loaded from different source systems and stored in the fractal data model which can dynamically support the change data. As a tool for enterprise information exploration, Oracle Endeca information discover is used for advanced, intuitive exploration and correlation analysis of these multidimensional multivariate data.

"With comprehensive, soft and hard integration, high-performance and cost-effective outstanding features, at present, Oracle large data solutions have been supported by many enterprise users." "Sishing said.