By Neal Singer
Sandia and supercomputer manufacturer Cray Inc. have signed a cooperative research and development agreement (CRADA) to form an institute focused on data-intensive supercomputers.
The Supercomputing Institute for Learning and Knowledge Systems (SILKS), to be located at Sandia/New Mexico, is expected to leverage the strengths of both Sandia and Cray by making software and hardware resources available to researchers who focus on a relatively new application of supercomputing. The task of such supercomputers is to make sense of huge collections of data rather than the traditional modeling and simulation of scientific problems.
“It’s an unusual opportunity,” says Bruce Hendrickson, Sandia senior manager of computational sciences and math (1440). “Cray has an exciting machine [the XMT] and we know how to use it well. This CRADA should help originate new technologies for efficiently analyzing large data sets. New capabilities will be applicable to Sandia’s fundamental science and mission work.”
Shoaib Mufti, director of knowledge management in Cray’s custom engineering group, says, “Sandia is a leading national lab with strong expertise in areas of data analysis. The concept of big data in the HPC [high-performing computing] environment is an important area of focus for Cray, and we are excited about the prospect of new solutions that may result from this collaborative effort with Sandia.”
Says Rob Leland, director of computing research (1400), “This is a great example of how Sandia engages our industrial partners. The XMT was originally developed at Sandia’s suggestion. It combined an older processor technology Cray had developed with the Red Storm infrastructure we jointly designed, giving birth to a new class of machines. That’s now come full circle. The institute will help leverage this technology to help us in our national security mission work, benefiting the Labs and the nation as well as our partner.”
The XMT has a different mode of operation from conventional parallel-processing systems. Says Bruce, “Think about your desktop: The memory system’s main job is to keep the processor fed. It achieves this through a complex hierarchy of intermediate memory caches that stage data that might be needed soon. The XMT does away with this hierarchy. Though its memory accesses are distant and time-consuming to reach, the processor keeps busy by finding something else to do in the meantime.”
In a desktop machine or ordinary supercomputer, Bruce says, high performance can only be achieved if the memory hierarchy is successful at getting data to the processor fast enough. But for many important applications, this isn’t possible and so processors are idle for most of the time. Said another way, traditional machines try to avoid latency (waiting for data) though the use of complex memory hierarchies.
XMT embraces latency
The XMT doesn’t avoid latency; instead, it embraces it. By supporting many fine-grained snippets of a program called “threads,” the processor switches to a new thread when a memory access would otherwise make it wait for data.
“Traditional machines are pretty good for many science applications, but the XMT’s latency tolerance is a superior approach for lots of complex data applications,” Bruce says. “For example, following a chain of data links to draw some inference totally trashes memory locality because the data may be anywhere.”
More broadly, he says, the XMT is very good at working with large data collections that can be represented as graphs. Such computations appear in biology, law enforcement, business intelligence, and in various national security applications. Instead of a single answer, results are often best viewed as graphs. Sandia and other labs have already built software to run graph algorithms, though “the software is still pretty immature,” Bruce says. “That’s one reason for the institute. As semantic database technology grows in popularity, these kinds of applications may become ubiquitous.”
Among its other virtues, the XMT saves power because it runs at slower speeds.
SILKS’ primary objectives, as described in the CRADA, are to accelerate the development of high-performance computing, overcome barriers to implementation, and apply new technologies to enable discovery and innovation in science, engineering, and for homeland security. The CRADA’s main technical categories include software, hardware, services, outreach, education, and training.
University students and faculty, as well as scientists and engineers from industry and government, are expected to be invited to take part in and benefit from the institute’s research. CRADAs are written agreements between a private company and a government agency to work together on a project.
A CRADA allows the federal government and non-federal partners to optimize their resources, share technical expertise in a protected environment, share intellectual property emerging from the effort, and speed the commercialization of federally developed technology. - Neal Singer
With more than 137 million artifacts, the Smithsonian Institution is the world’s largest museum and research complex. It comprises a mind-boggling scope of treasures, representing America’s rich heritage, art from around the globe, and the immense diversity of the natural and cultural world.
This month, nine historically significant robots from Sandia are joining the collection, where they will be permanently housed in the National Museum of American History, home of more than 3.3 million pieces of US history, including the wool and cotton flag that inspired the Star Spangled Banner, Kermit the Frog, and the desk Thomas Jefferson used to draft the Declaration of Independence.
“For the Smithsonian to request Sandia technology to be in their collections is an external recognition of the significance of Sandia National Laboratories’ contributions to the nation,” says Philip Heermann (6530) senior manager of Intelligent Systems, Robotics and Cybernetics and participant in the signing ceremony at the museum. “These robots will be in the same collections as some of Thomas Edison’s first electric light bulbs and Samuel Morse’s original experimental telegraph. The Smithsonian selected Sandia robots for inclusion after they researched the history of robotics and they found worldwide references, all pointing back to Sandia Robotics as early pioneers. The Sandia robots are similar to Edison’s electric light bulb in that both are first steps and testaments to American innovation.”
MARV, the tiny marvel
A Smithsonian curator contacted Sandian and former robotics engineer Ray Byrne (5535) earlier this year about obtaining some of the MARV, or Miniature Autonomous Robotic Vehicles, that made headlines in the mid-1990s as one of the first miniature robots developed in the US. Taking up no more than one cubic inch of space, MARV housed all necessary power, sensors, computers, and controls on board. Such an accomplishment held promise for exciting future developments and applications for medicine and the military.
Retired Sandia robotic senior scientist Barry Spletzer, who was instrumental in creating MARV and the Hopper, spoke at the transfer ceremony about the significance of the robots.
“Nothing like MARV had ever been built before,” Spletzer says. “We never expected recognition and certainly never thought we’d end up in the Smithsonian. This is certainly a career achievement.”
As the Smithsonian soon found out, MARV was just the tip of the iceberg of Sandia’s contributions to the advancement of the robotics field.
“The curator said they were looking for anything of historical significance. We have a lot that fit that requirement, so I started mentioning all of these older robots, and she was very interested,” Ray says. “So far, we’ve donated Dixie, the first battlefield scout robot, SIR, one of the first truly autonomous interior robots, the hopping robots, the NETBOTS, MARV, and the descendants of MARV, the super-miniature robots.”
The Sandia Interior Robot, or SIR, made a lasting impression on the nation when it was introduced in 1985 as the first truly autonomous interior robot. At the time, SIR was the only robot able to navigate a building without a preprogrammed pathway or floor wiring to find its way. It could run in manual or autonomous modes using navigational software, also developed at Sandia. SIR could perform dangerous work, such as disposing of radioactive waste or reconnaissance in a hostile environment.
In 1987, Sandia unveiled Dixie, the first battlefield scout robot. The all-terrain vehicle could perform reconnaissance work and exploration missions in a variety of landscapes. Dixie uses teleoperation with advanced navigation aides to enhance a remote operator’s understanding of surrounding terrain. The Hopper made news when it debuted in 2000 for its unique ability to navigate over walls and other obstacles by hopping 20 feet in the air over them.
With applications for planetary exploration, gathering war-fighting intelligence, and assisting police during standoffs or surveillance operations, the Hopper was the first robot powered by a combustion cylinder and a piston foot, and the wheeled Hopper was the first hybrid hopping/wheeled mobility system.
Sandia continued to wow the robotics field with the introduction of “superminature robots” in 2001. These tiny robots descended from MARV and were built small enough to be able to scramble through pipes or buildings to look for human movement or chemical plumes. Less than a quarter-cubic-inch in size, these robots could “turn on a dime and park on a nickel” and could include such enhancements as a miniature camera, microphone, communication device, and chemical microsensor. They had the ability to communicate with one another and work together, much like insects in a swarm. The superminiature robots were selected by Time as the invention of the year in robotics in 2001.
The related NETBOTS are roughly the size of a remote-controlled toy car, and in fact, built on the same platform. With more than 20 vehicles in the group, at the time they comprised the largest team of cooperating small robots ever developed. They could communicate and localize with respect to one another. NETBOTS operated on a network that allowed vehicles to pass messages and camera images to other vehicles out of the line of sight, and had applications for military and explosive ordinance removal. Ray took all of the robots on the plane, either in carry-on or checked luggage, to Washington, D.C., for a transfer ceremony to kick off the museum’s festivities for National Robotics Week.
“These are historically significant,” Ray says. “I am pleased that the Smithsonian has chosen to recognize Sandia’s contribution to robotics.” -- Stephanie Hobby
Jose Zayas (6120) and Dale Berg (6121) have been honored by Windpower Engineering magazine as two of the nation’s innovators and influencers in wind energy. Jose, senior manager of the Renewable Energy Technologies group, was named an influencer on wind energy. The magazine also named Berg, principal member of the technical staff at Sandia, an innovator of wind energy technology.
Jose sets priorities for Sandia’s Renewable Energy Technologies group at Sandia, and wind is an important part of that portfolio. With more than 14 years of experience in wind energy, Jose seeks ways to expand and accelerate Sandia’s role in the innovation, development, and use of all renewable energy technologies.
Jose recently led development of advanced water power systems, focusing on the emerging clean energy portfolio of wave, current, tide, and conventional hydro energy sources. He also leads a federal interagency research effort to overcome barriers to the continued deployment and acceptance of wind energy systems nationwide.
Jose joined Sandia in 1996 as a senior member of the technical staff. He holds a bachelor’s degree in mechanical engineering from the University of New Mexico and a master’s degree in mechanical and aeronautical engineering from the University of California.
Dale has worked throughout his career on key innovations that make wind systems reliable and competitive sources of energy. In the early 1980s, he helped develop the first airfoils designed specifically for wind turbine applications. Up to that time, blade airfoils were the same as those used on aircraft and sailplanes.
In the late 1980s, Dale contributed to the aerodynamic and structural design of the technologically innovative variable-speed Sandia 34-meter Testbed Vertical Axis wind turbine. More recently, Dale worked on turbine aerodynamics and the development of systems for aeroacoustics testing, analysis, and data acquisition. Aero-acoustics testing pinpoints the sources of noise generated by turbine blades with the aim of developing quieter blades.
Dale leads a multidisciplinary team of Sandia employees and contractors developing a turbine rotor that will reduce turbine damage due to frequent wind variations. The rotor integrates blade-mounted load and flow sensors, small, fast-response blade control surfaces, and embedded intelligent control systems, which are frequently referred to as “smart” rotors.
Dale has worked in wind energy at Sandia since 1981. He holds a bachelor’s degree from Michigan State University, a master’s in mechanical engineering from the University of New Mexico, and a doctorate in aeronautics from the California Institute of Technology.
The special Influencers and Innovators section of Windpower Engineering is available online.
Sandia got involved in the wind industry during the 1979 oil embargo. Dixie Lee Ray, head of the Atomic Energy Commission, challenged the national laboratories to examine alternative energy sources to decrease the nation’s dependence on foreign oil. In the early days, wind energy researchers at Sandia were referred to as “wind weenies.”
The industry was young and full of long-haired engineers/inventors working on designs in their back yards and garages, says Dale Berg (6121) Berg says then-manager Randy Maydew made contacts in Canada with other researchers working on so-called Darrieus vertical axis wind turbines, which were often referred to as “egg beaters.”
Randy took a staff member with him to the Natural Resources Council. He returned intending to investigate vertical axis wind turbines (VAWTs). His group mounted a small turbine on the roof of Bldg. 802 with great fanfare, but quickly realized that the dynamics of the blades meant that a less-traveled location would be safer. They then established a test bed southeast of the current Tech Area 1 off Poleline Road in the mid-1970s, and VAWTs remained there until the early 1990s.
The largest units were 17 meters in diameter. The vertical axis designs eventually gave way to the horizontal turbine designs common today, in part because forces were easier to understand, and market decisions eventually led to the end of research on vertical axis turbines.
Since then, the horizontal axis turbines have taken over and research has largely concentrated in that area. But old ideas find their way back in science; researchers are re-examining those vertical-axis designs for very large turbines, in part because gravitational loads on the blades are constant on the vertical axis machines, in contrast to the oscillating gravitation loads on the blades on the horizontal axis machines. The “egg beaters” may rise again to power our future. -- Stephanie Holinka