New IBM Servers with Tesla P100 GPUs and NVLink Mark a Milestone in High Performance Computing
Data center workloads are changing. Not long ago these systems were primarily used to handle storage and serve up web pages, but now they’re increasingly tasked with AI workloads like understanding speech, text, images and video or analyzing big data for insights.
Billions of consumers want instant answers to a multitude of questions, while enterprise companies want to analyze mountains of data to better serve their customers’ needs. Where do those answers come from? Data centers.
As a leader in server systems, IBM saw this trend coming several years ago, and partnered with us to accelerate new data center workloads. After four years of development, IBM today introduced its Power System S822LC for High Performance Computing powered by NVIDIA Tesla P100 GPUs and NVLink to facilitate high-performance analytics and enable deep learning on ever increasing mountains of data.
The system couples two of IBM’s POWER8 CPUs with four NVIDIA Tesla P100 GPUs connected via our NVLink high-speed interface. This is a custom-built GPU accelerator server, where the NVLink interface is routed on the motherboard and uses our Tesla P100 SXM2 GPU.
This tight coupling of IBM and NVIDIA technology enables data to flow 5x faster than over PCIe, accelerating time to insight for many of today’s most critical applications, like advanced analytics, deep learning and AI.
“The user insights and the business value you can deliver with advanced analytics, machine learning and artificial intelligence is increasingly gated by performance,” says Doug Balog, general manager of IBM Power Systems. “Accelerated computing that can really drive big data workloads will become foundational in the cognitive era. Based on OpenPOWER innovations from partners such as NVIDIA, our new OpenPOWER Linux servers with POWERAccel set a new standard for these workloads.”
The Path to Summit and Sierra
IBM has already lined up several customers, including a large multinational corporation and a number of research organizations, including the U.S. Department of Energy’s Oak Ridge National Laboratory (ORNL) and Lawrence Livermore National Laboratory (LLNL).
ORNL and LLNL will use the new systems as a development platform for optimizing applications to take advantage of NVLink. The systems will serve as a test bed for developing applications for Summit and Sierra, the next-generation supercomputers that IBM expects to deliver to ORNL and LLNL in 2017.
“The NVLink technology on the POWER platform provides coherency among the multiple memory hierarchies in the CPUs and GPUs,” said Arthur S. (Buddy) Bland, Oak Ridge Leadership Computing Facility project director at ORNL. “As a long-time user of GPUs, we believe that this will improve the performance of our applications and make it easier for the users to deliver great science.”