- IBM announced a new supercomputing server for high-performance cognitive computing applications, AI and ML.
- The new server, already used by large US labs and a large undisclosed multinational retail corporation, is powered by NVIDIA technology.
- The tight coupling of IBM and NVIDIA technology accelerates critical applications such as advanced analytics and deep learning for Big Data.
International Business Machines (NYSE:IBM) announced three new servers built for cognitive computing, Artificial Intelligence (AI) and Machine Learning (ML). These are three buzzwords for essentially the same thing: computer systems that learn how to solve more and more complex real-world problems and understand the world like humans. The automated machine learning technology of Watson, IBM's cognitive computing software platform, is at the forefront of AI research, but there are practical spin-offs that can already be used in advanced medical applications and commercial applications built on the Watson Developer Cloud.
The new servers are part of IBM’s Linux specialized server portfolio co-developed with members of the OpenPOWER Foundation. The OpenPOWER ecosystem includes a wide variety of computing solutions that use IBM's open POWER processor technology. In September 2015, Amigobulls covered IBM's plans of advance toward exascale supercomputing based on its POWER technology.
“The user insights and the business value you can deliver with advanced analytics, machine learning and artificial intelligence is increasingly gated by performance," said Doug Balog, General Manager of POWER, IBM 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."
One of the new servers, the Power System S822LC for High-Performance Computing, leverages the power of NVIDIA's (NSDQ:NVDA) Tesla P100 Graphical Processing Units (GPUs) and NVLink to deliver high-performance analytics and enable deep learning applications for Big Data. In the Power System S822LC, two new IBM POWER8 processors are directly connected with four NVIDIA Tesla P100 Pascal GPUs through NVIDIA NVLink, a high-speed, energy-efficient bidirectional interconnect. NVIDIA NVLink is embedded at the silicon level and incorporated into the overall system design.
The tight coupling of IBM and NVIDIA technology enables five times faster internal data flow, accelerating critical applications such as advanced analytics, deep learning and AI. According to IBM, the GPUs can deliver 21 teraflops of half-precision floating point performance. "Given how long it takes to train a machine learning model, those numbers can quickly make a difference in real-world applications," notes TechCrunch.
"The open and collaborative model of the OpenPOWER Foundation has propelled system innovation forward in a major way with the launch of the IBM Power System S822LC for High Performance Computing," said Ian Buck, VP of Accelerated Computing at NVIDIA. "NVIDIA NVLink provides tight integration between the POWER CPU and NVIDIA Pascal GPUs and improved GPU-to-GPU link bandwidth to accelerate time to insight for many of today’s most critical applications like advanced analytics, deep learning and AI."
In an NVIDIA blog post that covers the IBM announcement, Buck notes that data centers are 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, and enterprises want to analyze mountains of data to better serve their customers' needs. "Where do those answers come from? Data centers," says Buck. "As a leader in server systems, IBM saw this trend coming several years ago, and partnered with us to accelerate new data center workloads."
Two supercomputing centers in the US, Department of Energy’s Oak Ridge National Laboratory (ORNL) and Lawrence Livermore National Laboratory (LLNL), will use the new IBM server as a development platform for optimizing applications to take advantage of the built-in NVIDIA NVLink interface technology. "The combination of IBM POWER8 and NVIDIA Pascal via NVLINK will accelerate our mission-critical applications,” said Bronis R. de Supinski, CTO of Livermore Computing (LC) at LLNL. Arthur S. (Buddy) Bland, OLCF Project Director at ORNL, added that the NVLink technology on the POWER platform will improve the performance of ORNL applications and make it easier to deliver great science.
Federal laboratories and science are useful for image-building and public relations, but of course, the money is in the commercial application. IBM announced that a large undisclosed multinational retail corporation is among the early users of the new servers.
This is very good news for IBM investors, but even better news for NVIDIA investors. In fact, while large tech companies like IBM compete to deliver faster servers and AI applications for Big Data, NVIDIA is pursuing a pick-and-shovel strategy - selling key components to all the participants in the race - in the niche of specialized hardware accelerators for cognitive computing and AI, where it is the leader and has only a few competitors. So, NVIDIA can sell critical subsystems to all tech giants, as shown by the recent partnership with Baidu to create a cloud-to-car autonomous car platform for self-driving car makers, and confirmed by the spectacular performance of NVIDIA stock in the past year. The obvious advice to tech investors is, buy NVIDIA stock now.