Open source Duel, MapR Apache drill into enterprise applications

"Editor's note" Recently, MAPR has formally integrated the Apache drill into the company's large data-processing platform, and opened up a series of large database-related tools. Today, in the highly competitive field of Hadoop, open source has become a tool for many companies, they have to contribute more code to protect themselves, but also through open source to attack other companies. In this case, Derrick Harris made a brief analysis on Gigaom.

The following translation

Recently, the founder of the Mapr,apache Drill project has integrated the initial version of the technology into the company's large data platform. The company says this version of drill is 0.5, showing the SQL query engine in a "developer Preview" format.

Drill was first announced in August 2012, focusing on SQL on Hadoop, which has made great strides now. In the area of SQL on Hadoop, companies are the Impala, Hortonworks iterations of eight recount--cloudera, and a variety of start-ups and open source projects, including the immediate hive community.

However, MapR's figuratively Marketing officer Jack Norris said that drill is a technology worth looking for, because it is the "parent set" of all the features of other SQL on Hadoop engines. At the same time, the main feature of drill is that it can quickly generate structural patterns before the data is loaded into the database, primarily because instead of converting the data into other schemas or tables, the drill retains its original format. As a result, drill also fails to meet the needs of those who expect to convert data to a specific format.

Tomer Shiran, head of MAPR product management, said: "We are more confident that we can do a better job of drill than other SQL on Hadoop projects." ”

Although the company's current large data platform has integrated drill, the technology is not the only option for MAPR products. MAPR's large data platform also consolidates the Impala and hive stacks, and even supports HP's Vertica analysis tools with tighter integration.

Tomer admits: "By supporting more technology and contributing a lot of code, this will be part of a broader strategy for MAPR to reshape proprietary Hadoop vendors." Norris said:

At the moment, all the application-layer components in the MAPR release use open source technology or standard APIs. In the future, MAPR will be as open as possible to more technology. MAPR will confirm this with practical actions, such as this Tuesday, when the company Open-source a large number of Hadoop platform resource management capabilities, and submitted a MAPR disk IO allocation method, as well as job scheduling mechanism to Apache.

Drill has been supported and contributed by more than 40 companies, including Cisco, LinkedIn and the University of Wisconsin. Now, open source has been proven to be an effective way to improve the product, through the crowdsourcing way it can greet a large number of engineers dedicated to open source. At the same time, in the highly competitive field of Hadoop, open source is the sword that defends its own shield and attacks other companies on an open hand.

Original link: Sql-on-hadoop Tech Apache Drill is ready to use and part of MAPR ' s distro (Compile/Zhonghao revisers/wei)

Free Subscription "CSDN cloud Computing (left) and csdn large data (right)" micro-letter public number, real-time grasp of first-hand cloud news, to understand the latest big data progress!

CSDN publishes related cloud computing information, such as virtualization, Docker, OpenStack, Cloudstack, and data centers, sharing Hadoop, Spark, Nosql/newsql, HBase, Impala, memory calculations, stream computing, Machine learning and intelligent algorithms and other related large data views, providing cloud computing and large data technology, platform, practice and industry information services.