AI and Analytics at the Speed of Supercomputing
Cray's Urika®-XC AI and big data analytics tools bring you advanced analytics, machine learning and graph tools, engineered for Cray XC series supercomputers. Do It All
Gain new insights and solve the toughest problems across all your data on one machine — without additional hardware and without moving data from machine to machine. The Cray Urika-XC suite allows you to run big data analytics tools, machine learning and deep learning applications on the same system at the same time, on a system also running high-performance simulations and using familiar system tools and schedulers. With supercomputer speed and efficiency, you can run more types of analytics and artificial intelligence workloads faster, yielding unprecedented breadth of results and maximizing your opportunity for transformative discoveries.
Use modern machine learning frameworks, big data analytics tools, and data science languages to power richer insights and reduce the need for scarce talent. Cray’s Urika-XC suite includes a set of powerful big data analytics, data science and AI tools:
- Big data analytics toolswith Apache® Spark™ and the Spark ecosystem running at supercomputer scale and pbdR (Programming with Big Data using R)
- Data science tools: Python-based tools and libraries for data scientist productivity (Anaconda and Distributed Dask) and Jupyter notebooks for end-user productivity.
- Machine and Deep learning frameworks (TensorFlow™, PyTorch and Keras) and libraries (hyperparameter optimization and distributed training) enhanced to run at supercomputer scale (using 1,000 CPUs or GPUs) with simplified setup and administration.
- Graph databases using Cray’s powerful Cray Graph Engine (CGE) for big data graph pattern recognition and discovery applications. For the most demanding graph problems where size and performance matter, CGE has been tested against a 400-billion-tuple collection and shows a nearly 100x performance speed-up over Spark-based GraphX.
Innovate with Deep Learning Applications at Supercomputer Scale
You can capitalize on the XC series supercomputer’s scale and performance for your deep learning research with the Cray-developed library for distributed training (Cray PE ML plugin) and hyperparameter optimization (HPO). The Cray PE ML plugin is delivered in conjunction with popular machine intelligence frameworks (TensorFlow™, PyTorch and Keras) and leverages the Aries™ interconnect to give you scaling to more than 500 nodes on both CPUs and GPUs. Cray’s HPO libraries eliminate much of the guesswork required by a data scientist to set up a complex model and reduces the time required to achieve model accuracy.
With Urika-XC advanced analytics, you can make the most of your investment in Cray XC series supercomputers and reduce the need to maintain separate systems. Meet the increased demand for advanced analytics without capital investment or retraining.