IBM today announced it is bringing its Data Science Experience and PowerAI together to offer machine learning and deep learning on a single machine.

The Data Science Experience gives users collaboration tools for managing and monitoring data models, according to Dinesh Nirmal, IBM’s vice president of analytics development. PowerAIi, meanwhile, brings in GPUs as well as deep learning libraries and algorithms that can be used on multiple frameworks, such as TensorFlow, he said. By integrating PowerAI into the Data Science Experience, users will be able to create and train AI-based models using the deep learning frameworks to gain new insights into the ever-growing mountain of data being generated in today’s world.

“Data is the next natural resource,” Nirmal said. “Just [Internet of Things] data alone is 80 terabytes a day. There are massive amounts of data to process.”

Nirmal said that 80% of enterprise problems can be solved with machine learning, but deep learning is more effective for specific use cases. “If you’re running a huge neural network, that complexity requires deep learning. Or if you’re FedEx, to know what happened to a damaged box and how it got damaged, you would use deep learning. Anything that is data and process intensive,” he said.

IBM recently rolled out the Distributed Deep Learning library in PowerAI, created to reduce the time it takes to train “from weeks to hours,” according to a company statement announcing the PowerAI-DSX integration.

About David Rubinstein

David Rubinstein is editor-in-chief of SD Times.