Nvidia's Tesla K80 GPUs are helping IBM Watson rev up its cognitive computing capabilities.
IBM today announced it is using Nvidia Tesla K80 graphics processing units (GPUs) to accelerate the retrieve and rank API capabilities of its Watson cognitive computing system.
The announcement comes amid news that IBM and several fellow OpenPower Foundation members today introduced new technologies, collaborations and developer resources to enable organizations to analyze data more deeply and with more speed. IBM made its announcements at an IBM and OpenPower Accelerating Innovation event in Austin, Texas, where they previewed the next iteration of the Watson cognitive computing system architecture.
IBM said that by using the Nvidia Tesla K80 GPUs—the flagship offering of the Nvidia Tesla Computing Platform—coupled with Watson's Power-based architecture, Watson's retrieve and rank API capabilities increased by 1.7 times its normal speed. This speed-up can further improve the cost/performance of Watson's cloud-based services, IBM said.
"For example, if a call center agent is responding to an individual's health and insurance query, the agent will be able to leverage Watson's natural language processing technology to obtain an answer in real-time even faster and cheaper than before," said Rob High, an IBM Fellow and CTO of Watson, in a blog post. In addition to bolstering response time, the GPU acceleration also increases Watson's processing power to 10 times its prior performance, IBM said.
High said Watson often requires a lot of compute power and time to digest, annotate and index large quantities of data to prepare it to perform its cognitive tasks. Nvidia's Tesla GPUs will accelerate the time it takes the cognitive computing system to process this information to enable it to interact in natural language. Moreover, the combination of IBM Power architecture and Nvidia's Tesla platform will also facilitate the expansion of Watson's deep learning functionality. Hardware acceleration is integral to Watson's ability to deeply reason about vision and speech recognition—in a short period of time, he noted.
According to Nvidia, the Tesla platform has grown steadily since 2008 in the number of supported scientific, engineering, data analytics and other applications, with 370 GPU-accelerated applications now available.
A new study by Intersect360 Research, a technology research firm, shows that nearly 70 percent of the 50 most widely used high-performance computing (HPC) applications—and 90 percent of the top 10—support GPU accelerated computing. Among them are the ANSYS Fluent computational fluid dynamics application; the GROMACS molecular dynamics application; and now—announced today—VASP, an atomistic simulation application used by researchers around the world to model the behavior of individual atoms at the electronic level.
One of the study's authors, Addison Snell, CEO of Intersect360 Research, said: "Accelerated computing has reached the tipping point in HPC, with Nvidia's Tesla GPUs as the leader in the market. The adoption of accelerators and availability of GPU-accelerated versions of top HPC codes have been steadily increasing."
Indeed, "One day, all supercomputers will be accelerated," said Jen-Hsun Huang, co-founder and CEO at Nvidia, in a statement. "Leading supercomputing sites around the world have turned to GPU-accelerated computing, reflected in today's Top500 list. As the pace of discovery accelerates and researchers turn to computation, machine learning and visualization, we fully expect to see this trend increase."
Also today, IBM and newest OpenPower Platinum member Xilinx announced a multiyear strategic collaboration to jointly develop data center and network function virtualization (NFV) solutions that bring together the systems, software and management components around Xilinx field-programmable gate array (FPGA) accelerators.
In addition, two OpenPower members, E4 Computer Engineering and Penguin Computing, introduced new systems based on the OpenPower design concept incorporating IBM POWER8 and Nvidia Tesla GPU accelerators.
Also, responding to growing demand from developers for test servers to build, port and optimize applications that can take advantage of accelerators on Power-based systems, IBM and fellow OpenPower members have built a global network of physical centers and cloud-based services for no-charge access to accelerated Power-based infrastructure. New and expanded resources include expanded GPU services on SuperVessel. Nvidia and IBM worked together to accelerate SuperVessel, a cloud-based OpenPower ecosystem resource launched in June. SuperVessel now provides GPU-accelerated computing as-a-service capabilities, giving users access to high-performance Nvidia Tesla GPUs to enable Caffe, Torch and Theano deep-learning frameworks to instantaneously launch from the SuperVessel cloud.
And last month a team of noted researchers led by Erez Lieberman Aiden, a geneticist and computer scientist with appointments at Baylor College, Baylor College of Medicine and Rice University, was nationally recognized for breakthrough genomics research—a new procedure designed to modify how a human genome is arranged in three dimensions in the nucleus of a cell, with extraordinary precision.
Today, the team explained that the feat was achieved using Power Systems accelerated with Nvidia Tesla GPUs and Mellanox network infrastructure to build a 3D map of the human genome and model the reaction of the genome to this surgical procedure, without disturbing the surrounding DNA.