Cloudera and NetApp announced partnerships in Monday

Cloudera and NetApp announced a partnership in Monday, and NetApp will sell Cloudera's Apache Hadoop distribution and enterprise management software as agreed by both parties Cloudera will support the storage baseline open Solution for Hadoop that NetApp is about to release in December. This collaboration is clearly a response to NetApp's collaboration with MAPR Technologies in May this year.

As part of that deal, EMC entered the Hadoop Enterprise support business, competing directly with Cloudera, and integrating MAPR software into the Greenplum HD Enterprise Edition Hadoop software release. Hadoop, a distributed data processing platform based on Java, has become more and more popular with enterprise users in recent years with its ability to handle massive amounts of information. Storage-area opportunities naturally attract the attention of storage vendors, with EMC earning $17 billion trillion, and its competitor NetApp earning around $5 billion a year.

Cloudera is the oldest and largest supplier of enterprise support and Hadoop management software, with more than 100 enterprise users, but its size is insignificant compared to NetApp. It intends to dramatically improve its product sales and performance with the help of NetApp's sales and distribution teams. According to Cloudera and NetApp, the open Solution for Hadoop benchmark architecture consisting of hardware and software will accelerate deployment and has passed performance testing in NetApp labs.

In addition, although the usual commodity servers used in the Hadoop architecture have limited flexibility in calculating capacity versus storage, Cloudera and NetApp say Open solution can separate storage and computation while providing greater efficiency and reliability. and enhance the management ability of enterprise environment. Jeffrey O ' Neal, senior director of NetApp Data center solutions, said: "In our tests, computing capacity can grow at the rate of application requirements, storage can grow at the speed of data requirements, and as customers have started to build more load, we think this solution can provide great benefits." ”

For example, the Hadoop node on a commodity server like a pizza box usually includes 8 hard drives, O ' Neal says, NetApp hardware can install 14 2TB of hard drives after a computer node, and is well prepared to ensure higher reliability. RAID storage is also built into data protection. Disk drives are configured on pallets because there are many spare disks, so the failed disks can be pulled out quickly and the nodes and servers will not be affected. In addition, this architecture provides NetApp NFS backup protection for specified nodes, which is a single point of protection in Hadoop deployments because the specified node controls all other nodes.

EMC's Hadoop solution will run on top of EMC Greenplum Modular Data Computing Appliance (DCA), and NetApp open solution is a key difference from it, O ' Neal says, open Solution does not force users to use special databases. In fact, DCA can also meet conventional data warehousing requirements through EMC's Greenplum database. "Cloudera supports a variety of database connections, including Teradata and Netezza as well as Oracle, MySQL and Vertica databases," said Ed Albanese, Cloudera's business development director. ”

IBM has released Biginsights software and support services based on Hadoop, Oracle and Microsoft have announced plans to increase their Hadoop distribution and support (and Oracle's Big Data appliance), This data-processing platform is clearly moving towards a wider range of applications. MAPR chief executive John Schroeder, in a statement, likened the Netapp-cloudera benchmark architecture to an application structure, saying that entering the Hadoop market by commercial vendors would help make the platform a safe choice for large data platforms. "Most businesses run Hadoop by installing software on commodity hardware, which is less than 100 dollars, so let's see how the market reacts to the Hadoop device," he said. ”

(Responsible editor: The good of the Legacy)