Deep learning stretches up to scientific supercomputers

August 14, 2018, US Department of Energy
Researchers delivered a 15-petaflop deep-learning software and ran it on Cori, a supercomputer at the National Energy Research Scientific Computing Center, a Department of Energy Office of Science user facility. Credit: Lawrence Berkeley National Laboratory

Machine learning, a form of artificial intelligence, enjoys unprecedented success in commercial applications. However, the use of machine learning in high performance computing for science has been limited. Why? Advanced machine learning tools weren't designed for big data sets, like those used to study stars and planets. A team from Intel, National Energy Research Scientific Computing Center (NERSC), and Stanford changed that situation. They developed the first 15-petaflop deep-learning software. They demonstrated its ability to handle large data sets via test runs on the Cori supercomputer.

Using machine learning techniques on supercomputers, scientists could extract insights from large, complex . Powerful instruments, such as accelerators, produce massive data sets. The new software could make the world's largest supercomputers able to fit such data into deep learning uses. The resulting insights could benefit Earth systems modeling, fusion energy, and astrophysics.

Machine learning techniques hold potential for enabling scientists to extract valuable insights from large, complex data sets being produced by accelerators, light sources, telescopes, and computer simulations. While these techniques have had great success in a variety of , their use in for science has been limited because existing tools were not designed to work with the terabyte- to petabyte-sized data sets found in many science domains.

To address this problem a collaboration among Intel, the National Energy Research Scientific Computing Center, and Stanford University has been working to solve problems that arise when using deep learning techniques, a form of machine learning, on terabyte and petabyte data sets. The team developed the first 15-petaflop deep-learning software. They demonstrated its scalability for data-intensive applications by executing a number of training runs using large scientific data sets. The runs used physics- and climate-based data sets on Cori, a supercomputer located at the National Energy Research Scientific Computing Center. They achieved a peak rate between 11.73 and 15.07 petaflops (single-precision) and an average sustained performance of 11.41 to 13.47 petaflops. (A petaflop is million billion calculations per second.)

Explore further: Artificial intelligence may put private data at risk

More information: Thorsten Kurth et al. Deep learning at 15PF, Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '17 (2017). DOI: 10.1145/3126908.3126916

Related Stories

Artificial intelligence may put private data at risk

August 3, 2018

Machine learning – a form of artificial intelligence in which computers use data to learn on their own – is rapidly growing and poised to transform the world. But current models are vulnerable to privacy leaks and other ...

IBM demonstrates new breakthrough in AI performance

March 22, 2018

IBM has demonstrated a new breakthrough in AI performance. By using machine learning on POWER9 with NVIDIA Tesla V100 GPUs, IBM technology can now predict the likelihood of a user clicking online advertisements 46x faster ...

Recommended for you

Team breaks world record for fast, accurate AI training

November 7, 2018

Researchers at Hong Kong Baptist University (HKBU) have partnered with a team from Tencent Machine Learning to create a new technique for training artificial intelligence (AI) machines faster than ever before while maintaining ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.