Intel, Cloudera Tout Year-Old Big Data Partnership

Over the past year, the vendors have been able to accelerate data encryption in Cloudera's Hadoop distribution without hurting system performance.


Intel and Cloudera over the past year have improved the encryption offload performance of Cloudera's Hadoop distribution by 2.5 times while minimizing the impact on system performance.

Intel officials during the launch May 5 of the company's new high-end Xeon E7 v3 server processor family touted the chips' capabilities in accelerating data analytics and other compute-intensive workloads with more cores and features that enhanced security and reliability.

During the event, Cloudera CEO Tom Reilly took the stage with Diane Bryant, senior vice president and general manager of Intel's Data Center Group, to talk about the gains the two companies have made since the chip maker ditched its own Hadoop distribution last year and invested $740 million in Cloudera.

Since partnering with Intel, Cloudera has issued four releases of its Hadoop distribution and has optimized it for the Intel Architecture, including embedding Intel's AES-NI security features, which has improved the security of enterprise data. The work with Intel engineers on the Hadoop distribution has allowed Cloudera to enable full database encryption with minimal impact—from 1 percent to 4 percent system overhead—on the hardware.

Hadoop jobs can run at a faster rate and additional Hadoop workloads can be run, according to Reilly.

"Customers before would encrypt their sensitive data," he said from the stage at the Intel event. "Now they can encrypt all of their data."

Cloudera also has been able to use in its distribution some of the functionality that Intel had put into its Hadoop distribution, and the two vendors are working with other companies to help accelerate new capabilities and integrations of enterprise software with Cloudera's Hadoop.

Intel and Cloudera also have seen momentum in businesses adopting Cloudera's distribution running on Intel processors. Those businesses include Caesars Entertainment, health care technology vendor Cerner and online dating site eHarmony. In addition, MasterCard is implementing the Cloudera distribution, which has received full Payment Card Industry (PCI) certification. In an interview with eWEEK before the Intel event, Reilly said the adoption by MasterCard is an indication of the high level of security in his company's Hadoop distribution.

"There's been so much talk about big data and the promise of big data," he said. "We now have many, many use cases."

The CEO said he expects the number of those use cases to grow as the demand for faster data analytics and greater data security increases, given the rise in the amount of data that's being generated and the promise that such trends as the Internet of things (IoT) will only increase that amount.

"Not only is [the amount of] data growing, but it's growing fast and it's time-sensitive," Reilly said, adding that it needs to be analyzed quickly to give businesses the best results.

The partnership with Intel also has helped Cloudera's bottom line. According to the two vendors, since Intel's investment in March 2014, Cloudera's annual recurring subscription software revenue grew 100 percent over the same period a year earlier, and the number of enterprise subscription software customers grew by more than 85 percent.

The company is looking to continue the growth this year by extending its global reach by opening new offices in such areas as China.