Enabling NASA's advanced modeling and simulation
[image-62]Supercomputing, also called high performance computing (HPC) or high-end computing (HEC), has been recognized by Congress as "vital to the nation's prosperity, national and economic security, industrial production, engineering, and scientific advancement." Aided by federal investments across many agencies, HPC has become a powerful and indispensable tool for the advancement of many national priorities and federal agency missions.
NASA has been a key participant in the advancement and use of supercomputing in support of its missions. Applications have included modeling fluid dynamics phenomena and full aerospace vehicles, Earth’s weather and climate, solar physics and the formation of the universe, and much more. All of these disciplines require high-fidelity numerical modeling of complex systems and processes, and detailed analysis and visualization of large-scale data, both enabled by supercomputing, to advance human knowledge and technology.
NASA's Ames Research Center became prominent in supercomputing with its operation of the ILLIAC IV — by far the most power supercomputer in the world at the time — starting in the early 1970s. Building on this expertise, the Numerical Aerodynamic Simulation Program was approved by the U.S. Congress in 1983 to provide leading-edge computational capabilities and to dramatically advance aerospace modeling. The program established a model of partnership between supercomputing operations and users that continues today with the agency’s premiere NASA Advanced Supercomputing (NAS) Facility at Ames.
NAS provides an integrated environment including world-class HEC resources and services customized to meet NASA's unique high-fidelity modeling and simulation needs for Earth and space science, space exploration and aeronautics, serving users across the country from NASA centers, academia and industry. To meet these needs, NAS continually expands its supercomputing and storage resources, including Pleiades, one of the fastest computers in the world. NAS also provides comprehensive user services—application support, large-scale data analysis and visualization, network support, and user environment customization—to broadly accelerate NASA's science and engineering activities, and thus to enhance the success of NASA's mission.
Featured example: Quantum computing experimentation
Can quantum computing solve previously unsolvable problems?
NASA, the Universities Space Research Association (USRA) and Google are partnering to assess the potential of quantum computers to perform calculations that are difficult or impossible using conventional supercomputers. A quantum computer is being installed in the NASA Advanced Supercomputing (NAS) facility at NASA’s Ames Research Center, and is expected to be operational in early fall 2013.
Quantum computing is based on quantum bits or qubits. Unlike traditional computers, in which bits must have a value of either zero or one, a qubit can represent a zero, a one, or both values simultaneously. For certain types of optimization problems, the quantum properties of a qubit allow researchers to effectively try all possible solutions at once, arriving at the best solution much more quickly.
Potential benefits to NASA include dramatically improved algorithms for optimization tasks in air traffic control, autonomy, robotics, navigation and communication, system diagnostics, pattern recognition, and mission planning and scheduling.
The research team at Google will focus on how the quantum system may help build more accurate models for everything from speech recognition, to web search, to protein folding. The collaboration allows the broader academic community, through USRA, to utilize a quantum computer and to work together with researchers at both NASA and Google. The computer is being leased by USRA from D-Wave Systems.
› Read more at the USRA quantum computing website
Featured example: Advances in rotorcraft computational fluid dynamics
How has the unprecedented accuracy of our advanced modeling techniques increased the passenger capacity on the V22 Osprey and Blackhawk helicopters?
[image-67]Rotorcraft such as helicopters and tiltrotor aircraft provide many useful civil and military functions, without requiring airports and runways. Modern rotorcraft designs offer significant improvements in vehicle performance, safety, and reduced environmental impact. However, the accurate simulation of rotorcraft air flow with computational fluid dynamics (CFD) continues to be a challenging problem. Rotor blades generate blade-tip vortices as they spin around and encounter the vortices of other blades, resulting in very complex air flows. Additionally, rotorcraft simulation is multidisciplinary and must take into account the interaction between rotor blade aerodynamics, blade flexibility and blade motions to achieve controlled steady flight.
Using ultra-high resolution simulations on Ames supercomputers, the chief hover performance parameter has been predicted within experimental error for the first time for V22 Osprey and Blackhawk helicopter rotors in hover over a wide range of conditions. With typical methods, this parameter is under-predicted by 2-6%, where each half of one percent translates into one fewer passenger the vehicle can carry.
The Ames simulations revealed tremendous detail of the blade vortices and turbulent flow that had never before been observed computationally or experimentally. The error in vortex core growth rate has been reduced from 60% to 4%. Solutions typically required 1-2 weeks on Ames’ Pleiades supercomputer with 5-24 hours of computational time for each simulated rotor revolution. These groundbreaking results will impact the next generation of rotorcraft analysis and design.
› Read more
Featured example: Space Launch System (SLS) modeling
How is Ames’ ongoing modeling and simulation making an impact on important design decisions for NASA’s next generation launch vehicle?
[image-83]Ames is performing computational fluid dynamics (CFD) simulations for SLS on analysis of structure, booster separation and forces on the engine hinges, and providing critical information for down-selection from several vehicle shapes. Early design for the SLS has been focused on maintaining stability and control of the vehicle during ascent, and structural integrity throughout the mission. In order to determine details of the environment around the vehicle in these phases before it’s built, CFD simulations were performed at select points over the ascent path with various conditions. This informed critical design decisions for SLS early in the design cycle when wind tunnel data was not yet available.
Ongoing modeling and simulation support includes characterizing the aerodynamic performance of the vehicle for a suitable ascent trajectory, determining the distributed aerodynamic forces along the vehicle for structural analysis and providing surface pressure signatures to assist in venting design for parts of the vehicle.
NASA’s Advanced Supercomputing facilities enable fast and efficient turnaround time for CFD simulation. Ames’ Pleiades supercomputer allows several hundred complex simulations to be completed in under a week. With detailed calculated information about the aerodynamic forces on the vehicle, engineers can improve the SLS design with modifications to the shape and structural components.
› Read more
Featured example: Enabling collaboration with NASA Earth Exchange (NEX)
- How does the NASA Earth Exchange provide the scientific community with the capability to process nearly 5 trillion pixels in only a few hours?
[image-99]NEX delivers a complete work environment in which users can explore and analyze large datasets, run modeling codes, collaborate on new or existing projects, and quickly share results among the Earth science communities. It utilizes Ames’ Pleiades supercomputing platform together with over one petabyte of satellite, climate and model datasets. The Pleiades architecture combined with massive data store and a high-speed network enables NEX to engage large scientific communities and provide them with capabilities to execute modeling and data analysis on a grand scale, which was not previously achievable by most scientists.
In one recent application of NEX, a team of researchers from NASA and South Dakota State University used the system to derive a new global vegetation information product from over 80,000 Landsat scenes in order to provide a dynamic view of changing vegetation over the course of the year at 30-meter resolution. The processing of nearly 5 trillion pixels took only few hours on NEX, providing an unprecedented monitoring capability at a global scale.
As the development of NEX continues, it strives to lower the barrier of entry to data- and compute-intensive science. NEX will provide a mechanism for continuous engagement among members of the global science communities to work together to address grand challenge problems in Earth sciences.
› Read more