H2O Brings Predictive Analytics to Hadoop

MOUNTAIN VIEW, Calif., Feb. 4 — H2O, the open source in-memory machine learning and predictive analytics company for big data, today announced that its flagship H2O product is available on the Intel Distribution for Apache Hadoop (Intel Distribution). Companies that use the Intel Distribution are now able to use open source H2O to run advanced algorithms on existing data stored in Hadoop clusters with no need for data transfers. By combining the power of H2O’s highly predictive algorithms with the high performance Intel Distribution, organizations can discover valuable insights up to 100x faster than alternatives.

The key to H2O’s interactive performance is its fast in-memory parallel processing. Cache oblivious implementations of algorithms over columnar compressed data delivers distributed machine learning algorithms at blazing speeds.

“Big data is transformative for enterprises,” said SriSatish Ambati, co-founder and CEO of H2O. “By offering the H2O product with the Intel® Distribution for Apache Hadoop, customers will achieve near-real time predictions and nano-second scoring to prevent credit card fraud, customer churn and better sales predictions.”

“We are constantly bringing improvements in the development of hardware and software to ensure the two work best together and deliver the best performance and features needed in the industry,” said Jobi George, Director Business Development and Strategy of Intel’s Datacenter Software Division. “With the H2O product available on the Intel® Distribution for Apache Hadoop, customers will now be able to bring fast in-memory predictive analytics that enable new discoveries and insights.”

H2O delivers parallel and distributed advanced algorithms on big data at speeds up to 100x faster than other predictive analytics providers and is easy to install and deploy in place on big Hadoop clusters. With a simple click, data models can be expressed into scoring engines ready for low latency production environments.

A vibrant community of data scientists, systems and language enthusiasts has built up quickly around a shared interest in H2O. H2O has sponsored or participated in 45 data science meet-ups over the past 9 months. For more information on upcoming H2O meet-ups, please visit http://0xdata.com/events/ or join the movement at https://github.com/0xdata/h2o.