H2O largesse in-memory predictive analytics en route for Hadoop
MOUNTAIN VIEW, USA: H2O, the open source in-memory machine learning and predictive analytics company intended for big data, stated that its flagship H2O product is accessible on the Intel Distribution for Apache Hadoop (Intel Distribution).
Companies that make use of the Intel Distribution are at the present capable of using open source H2O to run advanced algorithms on current data stored in Hadoop clusters with no requisite for data transfers. By way of combining the power of H2O’s highly predictive algorithms with the high performance Intel Distribution, organizations can determine valuable insights up to 100xs more rapidly than another course of action, that is.
The key to H2O’s interactive performance is its fast in-memory parallel processing. Cache oblivious implementations of algorithms over columnar compressed data powers distributed machine learning algorithms at intense speeds.
“Big data is transformative for enterprises,” reportedly said Sri Satish Ambati, co-founder and CEO of H2O. Volumnising further that – “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.”
H2O supplies parallel and distributed advanced algorithms on big data at speeds totaling 100xs quicker than other predictive analytics providers and is stress-free to be installed and arrayed in place on big Hadoop clusters. By means of a simple click, data models can be expressed into notching engines all set for low dormancy production hinterlands.
