New technology bridges Oracle, Hadoop, NoSQL data storage

The use of large data has been far less than the ability to collect large data, the main reason is that the current enterprise data mainly dispersed in different systems or organizations, the key to the big data strategy is to be able to more in-depth, richer mining all the data system of valuable information, so more accurate prediction of customer behavior, find business value, However, it is difficult to move this data to a separate data store, and security and regulatory issues are not guaranteed, Oracle Big Data SQL launched to solve the current challenges.

The following is the translation:

Discovering the needs of a business or organization for a data management architecture, Oracle launches Big Data SQL software to consolidate a wide variety of data sources, including Hadoop, NoSQL, and Oracle databases.

A complete solution is to combine Oracle's large data devices with big SQL, Cloudera Hadoop distributed and Oracle's own NoSQL database. At the start Oracle Big Data SQL only supports Apache hive and Hadoop File System, and other vendors need to migrate SQL relational databases to run on Hadoop.

Oracle provides a complete set of solutions for large data devices and big SQL

Oracle's big Data SQL product means that administrators do not have to learn other query languages when dealing with information in a relational database or Hadoop, says Neil Mendelson, Oracle's large analysis director.

We can use the Oracle SQL language that we have become accustomed to using-the complete SQL language, and we can now access any of these three data sources directly or any combination of them, he explains.

Similar tools are available in open source communities such as Stinger, allowing you to use SQL commands to query data in Hadoop, or to use CQL languages (Cassandra Query languages) designed to implement SQL commands on NoSQL systems.

The goal of creating this large data management system is to expect SQL queries to run on different data sources, and to enable enterprises or organizations to leverage existing technologies to maintain enterprise-level data security and manage sensitive information. Oracle says the technology's unique architecture and smart scan inherit from Oracle Exadata, while allowing Oracle Big Data SQL to query all forms of structured and unstructured data and minimize data movement.

This also facilitates the security features of Oracle databases, including organizing existing security policies to extend to Hadoop and NoSQL data.

Oracle's Dan McClary says product development has been going on for some time, and it goes beyond existing technology. He also said that big Data SQL and HDFs datanodes and yarn nodemanagers were able to co-resident, and that queries from new external tables were sent to these services to ensure direct path reading and data localization.

Cloudera, founder, chairman and chief strategist Mike Olson, said: "Running Cloudera software on large Oracle data devices is more cost-effective and faster than DIY cluster deployments." When Hadoop queries data, we have seen a strong customer demand for SQL. ”