The U.S. Department of Energy has awarded IBM contracts valued at $325 million to develop and deliver advanced “data-centric” supercomputing systems at Lawrence Livermore and Oak Ridge National Laboratories.
“This architecture is really part of a paradigm that addresses the big data challenge, one we hear about here at IBM all time - which we call data-centric computing. We believe the value of a supercomputer is not only tied to petaflops but also to the speed of insights. We solve this particular challenge working with the labs through an open ecosystem leveraging technologies with our partners at the OpenPower Foundation, NVIDIA and Mellanox,” said Tom Rosamilia, senior vice president, IBM Systems and Technology Group, during a webcast to announce the new supercomputing systems.
The computers will be deployed to advance innovation in science, engineering and national security.
According to IBM, big data accelerates the opportunity for new discovery while at the same time magnifying the challenge scientists face. The current approach to computing presumes a model of data repeatedly moving back and forth from storage to processor in order to analyze and access data insights, process that is unsustainable with the onslaught of big data because of the amount of time and energy that massive and frequent data movement entails.
"We believe the big data challenges can only be solved through open innovation. The classic computing technologies will obviously continue to evolve but at a rate far short of the rate at which data is growing. We see the growth on our networks, on disks, and on microprocessors are all improving but not at the rate required to keep pace with the growth of data,” said Rosamilia.
The goal of data-centric computing is to eliminate data movement as much as possible, Rosamilia said. “The combination of IBM’s organic system innovation and the contributions from the OpenPower Foundation members presents us with a pathway to drive this new era of data-centric computing. Through collaboration-based on openness we are accelerating the amount and variety of innovation incorporated into our technical computing solutions.”
IBM OpenPOWER Systems
IBM says the OpenPOWER-based systems at each laboratory are expected to offer five to 10 times better performance on commercial and high-performance computing applications compared to the current systems at the labs, and will be more than five times more energy efficient.
The “Sierra” supercomputer at Lawrence Livermore and “Summit” supercomputer at Oak Ridge will each have a peak performance in excess of 100 petaflops balanced with more than five petabytes of dynamic and flash memory to help accelerate the performance of data centric applications. The systems will be capable of moving data to the processor, when necessary, at more than 17 petabytes per second to speed time to insights.
The Sierra and Summit systems will be used for the most mission-critical applications and represent the next major phase in the U.S. Department of Energy’s scientific computing roadmap to exascale computing.
The incorporation of OpenPOWER technologies into a modular integrated system will enable Lawrence Livermore and Oak Ridge to customize the Sierra and Summit system configurations for their specific needs.
Working with IBM, NVIDIA developed the advanced NVIDIA NVLink interconnect technology, which will enable CPUs and GPUs to exchange data five to 12 times faster than they can today. NVIDIA NVLink will be integrated into IBM POWER CPUs and next-generation NVIDIA GPUs based on the NVIDIA Volta architecture, allowing Sierra and Summit to achieve unprecedented performance levels. With Mellanox, IBM is implementing a state-of-the-art interconnect that incorporates built-in intelligence, to improve data handling.