CUDA In Action Spotlights
Read the customer spotlights below and learn how scientists and industry professionals are leveraging CUDA-based GPU computing across a range of disciplines and scientific applications.
Harvard University: Cris Cecka
Dr. Cris Cecka is a research scientist and lecturer in the Institute for Applied Computational Science (IACS) at Harvard. He also performs research at the Massachusetts Institute of Technology. Previously, Cris was a graduate student in the Institute for Computational and Mathematical Engineering (ICME) at Stanford University with Prof. Eric Darve. His research focuses on computational mathematics, particularly for interdisciplinary applications in science and engineering. "My research with GPU computing has really changed the way I approach software and algorithm development," he comments.
Univ. of Delaware: Michela Taufer
Dr. Michela Taufer joined the University of Delaware in 2007, where she was promoted to associate professor with tenure in 2012. Her research interests include software applications and their advanced programmability in heterogeneous computing (i.e., multi-core platforms and GPUs); cloud computing and volunteer computing; and performance analysis, modeling and optimization of multi-scale applications. Michela comments: "My team's work is all about rethinking application algorithms to fit on the GPU architecture in order to get the most out of its computing power, while preserving the scientific accuracy of the simulations. This has resulted in many exciting achievements!"
MIT: Mark Bathe
Dr. Mark Bathe is an Associate Professor of Biological Engineering at the Massachusetts Institute of Technology. His lab focuses on in silico design and programming of synthetic nucleic acid scaffolds for engineering light-harvesting antennas, multi-enzyme cascades, cellular delivery vehicles, and fluorescent biomolecular probes, which he assays using innovative quantitative imaging techniques. Mark comments: "GPU computing plays a major enabling role in diverse aspects of our technology developments."
Baidu IDL: Ren Wu
Dr. Ren Wu is a distinguished scientist at Baidu's Institute of Deep Learning (IDL). He is widely known for his pioneering research in using GPUs to accelerate big data analytics as well as his contribution to large-scale clustering algorithms via the GPU. Ren is also known for his work in computer game theory, and was the first person to perform systematic computational research on Xiangqi (Chinese chess) endgames. Ren comments: "CUDA as a programming model has become very mature over the years and is perfect for training deep neural networks, especially for large models." He adds: "We will see many breakthroughs in the next five years. Heterogeneous computing will become mainstream. We will continue to see much higher performance and more power efficient chips."
IDSIA: Dan Ciresan
Dr. Dan Ciresan is a senior researcher at IDSIA in Switzerland and a pioneer in using CUDA for Deep Neural Networks (DNNs). His methods have won international competitions on topics such as recognizing handwritten Chinese characters and detecting mitosis in cancer histology images. IDSIA is a research institute for artificial intelligence, affiliated with the University of Lugano and the University of Applied Sciences of Southern Switzerland. Dan comments: "DNNs will have a huge impact in the bio-medical field. They will be used for developing new drugs (e.g. automatic screening), as well as for disease prevention and early detection (e.g. glaucoma, cancer). They will help improve our understanding of the human brain."
University of Sheffield: Paul Richmond
This week's Spotlight is on Dr. Paul Richmond, a Vice Chancellor's Research Fellow at the University of Sheffield (a CUDA Research Center). Paul's research interests relate to the simulation of complex systems and to parallel computer hardware. He is the developer of FLAME GPU. Paul comments: "One of the most exciting aspects of GPU-accelerated simulation is that simulations can often be run faster than real-time. For example, a pedestrian evacuation model can be matched to real-world conditions and an evacuation plan can be simulated, essentially looking into the future for potential danger or problems. This forms the basis of my current work using agent simulation techniques for prediction and decision making."
Carnegie Mellon University: Ian Lane
This week's Spotlight is on Dr. Ian Lane. Ian is an Assistant Research Professor at Carnegie Mellon University where he leads a speech and language processing research group based in Silicon Valley. He co-directs the CMU CUDA Center of Excellence with Dr. Kayvon Fatahalian. Dr. Lane and his team are actively conducting research in hardware-accelerated perceptual computing, including GPU-accelerated methods for speech recognition, image processing and natural language processing understanding. "Over the past two years we have specifically been focusing on heterogeneous CPU-GPU platforms," Ian comments. "One of the breakthroughs was the execution of the core speech recognition search and language model lookup steps in parallel on the GPU and CPU cores respectively. This allowed us to perform speech recognition with extremely large language models (over one billion parameters), with little degradation to the speed in which speech recognition was performed on the GPU."
Stanford University: Todd Martínez
This week's Spotlight is on Professor Todd Martínez of Stanford. Professor Martínez' research lies in the area of theoretical chemistry, emphasizing the development and application of new methods which accurately and efficiently capture quantum mechanical effects. Professor Martínez pioneered the use of GPU technology for computational chemistry, culminating in the TeraChem software package that uses GPUs for first principles molecular dynamics. He is a founder of the company PetaChem that distributes this software. He comments: "TeraChem was written for GPUs....The payoff from this was remarkable – TeraChem is up to two orders of magnitude faster than similar programs designed for CPUs. This is bringing us much closer to the holy grail of computer-aided molecular design." Learn about the Martinez Research Group at Stanford [http://mtzweb.stanford.edu/mtz/
Artefacto Estudio: Néstor Gómez
This week's Spotlight is on Néstor Gómez, CEO of Artefacto Estudio in Mexico City. Artefacto Estudio is a developer of interactive applications and games. Néstor and his team leverage GPUs in their Microsoft Kinect-based real-time "virtual shoe fitting" application. "Profiling is very good on CUDA and it allows us to find bottlenecks easily," comments Néstor. "One of our favorite tools is NVIDIA Nsight Visual Studio Edition, which enables us to debug our parallel code and profile the execution times of every part of the tracking algorithm."
Kavli Institute: Debbie Bard
Debbie Bard is a cosmologist at the Kavli Institute for Particle Astrophysics and Cosmology (KIPAC). KIPAC members work in the Physics and Applied Physics Departments at Stanford University and at the SLAC National Accelerator Laboratory. To handle the massive amounts of data involved in cosmological measurements, Debbie and her colleagues Matt Bellis (now an assistant professor at Siena College) and Mark Allen (now a data scientist at Chegg) teamed up to explore the potential of GPU computing and CUDA. They concluded that "GPUs are a useful tool for cosmological calculations, allowing calculations to be made one or two orders of magnitude faster." Their results were presented in a paper titled Cosmological Calculations on the GPU, which appeared earlier this year in Astronomy and Computing. http://www.sciencedirect.com/science/article/pii/S2213133712000030
Hologic: Diego Rivera
Diego Rivera is a senior software engineer at Hologic, Inc. Hologic develops diagnostic products and medical imaging systems with an emphasis on serving the healthcare needs of women throughout the world. "GPU computing has allowed us to process and reprocess images in real time. The impact of this is that there is no wait time added for screening and diagnostic results, which in turn minimizes the patient's anxiety," says Diego.
The Favreau Brothers
Cyrille Favreau and Christophe Favreau rely on GPU computing in different ways, with equally compelling results. Cyrille uses CUDA to pursue his interest in visualization technologies. Christophe, a professional photographer, is passionate about sailing and nature. GPUs help him produce beautiful work as he travels the globe. Cyrille comments: "In 2009, I discovered CUDA and that took me to a whole new world. I could see that massively parallel architectures were about to shake the foundations of traditional programming."
Freie Universität Berlin: Knut Reinert
This week's Spotlight is on Dr. Knut Reinert. Knut is a professor at Freie Universität in Berlin, Germany. He and his team focus on development of algorithms and data structures for analysis of biomedical mass data, and are the creators of SeqAn, an open source C++ library. Previously Knut was at Celera Genomics, where he worked on the Human Genome Project. Knut comments: "Bioinformatics and computation will play an increasingly important role in the biomedical and healthcare fields.... The next ten years will bring a paradigm shift in medicine."
UC San Francisco: Adam Gazzaley
This week's Spotlight is on Adam Gazzaley of UC San Francisco, where he is the founding director of the Neuroscience Imaging Center and an Associate Professor in Neurology, Physiology and Psychiatry. Adam comments: "We are working with a distributed team (UCSF/Stanford/UCSD and Eyevapor) to CUDA-enable EEG processing to increase the fidelity of real-time brain activity recordings. The goal is to more accurately represent the brain sources and neural networks, as well as to perform real-time artifact correction and mental state decoding. Where CUDA and the GPU really excel is with very intense computations that use large matrices. We generate that type of data when we're recording real-time brain activity across many electrodes."
Vrije Universiteit: Pierre Wahl
Pierre Wahl is a PhD student at Vrije Universiteit in Brussels, Belgium. As a member of the Brussels Photonics Team (B-PHOT), he designs energy-efficient optical interconnects and works closely with the NVIDIA Application Lab at the Forschungszentrum Jülich. Pierre used CUDA to develop Belgium-California Light Machine (B-CALM), a GPU-accelerated Finite Difference Time Domain (FDTD) Simulator for applications in photonics. Pierre comments: "Using our CPU FDTD code, we quickly ran into computational limitations. B-CALM was born by porting our FDTD code to GPUs using CUDA and we obtained an 80X speed-up."
Luna: Patrick Roye
Luna, Inc. is a pioneer in fiber-optic shape and position sensing. Its technology is being developed to integrate into systems which perform minimally invasive diagnostics, surgery and therapy, pinpointing the position and shape of an instrument inside the body. Patrick works on accelerating Luna's processing algorithms using GPUs. He and a team of engineers and scientists are developing a prototype system that uses CUDA to calculate the shape of a fiber-optic sensor in real-time. "If you enjoy optimizing code and making it run blazing fast, then GPU programming is for you," comments Patrick.
Princeton: Valerie Halyo
Valerie Halyo is an assistant professor of physics at Princeton. As an expert in the area of high energy physics, she is exploring fundamental questions about the nature of the universe, looking for the elementary particles that constitute matter and its interactions. GPUs and CUDA are accelerating her research. "A great algorithm design isn't very useful if it can't be implemented in an efficient way on modern parallel processors," she says. Her team is using NVIDIA Tesla C2075 and K20 GPUs.
San Diego Supercomputer Center: Yifeng Cui
Yifeng Cui is director of the High Performance GeoComputing Lab (HPGeoC) at the San Diego Supercomputer Center and adjunct professor at San Diego State University. The HPGeoC Lab was recently named a winner of the HPC Innovation Excellence Award by IDC for developing a highly-scalable computer code that promises to dramatically cut research times and energy costs in simulating seismic hazards. "Computing is changing more rapidly than ever before, which is particularly challenging for earthquake applications," notes Yifeng.
Texas A&M: Jon Rogers
Jon Rogers is director of Texas A&M's Helicopter and Unmanned Systems Lab, where he works on new technologies for autonomous systems. Jon is currently exploring new algorithms and sensing technologies to increase task complexity of robotic devices. His research encompasses nonlinear dynamics, robust control, and high-performance computing. Jon says: "We leverage CUDA primarily for parallel trajectory simulation, which means we have developed dynamic models that run within a GPU kernel. Launching thousands of threads means we can run numerous dynamic simulations at once. CUDA specifically has allowed us to take existing codes and port them to the GPU relatively quickly."
Siemens Medical: Ismayil Guracar
Ismayil Guracar is a Senior Key Expert at Siemens Medical Solutions USA. He and his colleagues have applied CUDA to the high-data-rate signal processing pipeline of the ACUSON SC2000™ ultrasound imaging system. In addition to replacing existing FPGA- and CPU-based processing functions, the team has been able to rapidly extend the capabilities of the system to achieve improved speckle reduction, information content and lesion conspicuity. Ismayil notes: "The FPGA place and route cycle took half a day. With CUDA, compilation takes just seconds. We can go from idea to prototype to product a lot faster."
MIT Lincoln Lab: Dylan Jackson
Dylan Jackson works at MIT Lincoln Laboratory in the area of advanced optical systems. Previous to joining MIT-LL, Dylan was a student at Boston University. His master's degree project explored the application of GPU computing to the LED-based Interferometric Reflectance Imaging Sensor (IRIS) system. He says: "Since CUDA enables parallel computing on GPUs, it seemed to be an ideal platform for exploiting data level parallelism on a single computer. I developed a CUDA implementation of the IRIS model, yielding significant performance improvements over the current implementation."
University of Waterloo: Aron Broom
Aron Broom is a researcher at the University of Waterloo in Ontario, Canada, where he works on the molecular engineering of protein structure and function. Aron's research interests include protein-ligand binding (particularly of a multivalent nature) and stability and symmetry in protein structure and evolution. He comments: "Because of the incredible performance to price and performance to power ratios of GPUs, I'm able to study problems using small GPU clusters which would not be possible using larger traditional CPU-only clusters."
NTNU: Anne Elster
Anne C. Elster is an Associate Professor at the Norwegian University of Science and Technology (NTNU), a CUDA Research Center and CUDA Teaching Center, where she runs the HPC-Lab (Heterogeneous and Parallel Computing Lab). She is also a Visiting Scientist at the University of Texas at Austin, a CUDA Teaching Center. She comments: "It is very rewarding to see young minds get excited about technology. I feel it is important for our computer science students to learn about compiler and parallel programming given how fast the industry is moving with multi- and many-core systems, in everything from supercomputers to workstations to handheld devices."
nCore Design: Ian Lintault
Ian Lintault is managing director at nCore Design, a firm focused on applications that require high compute performance and low latency in embedded form factors. Ian comments: "GPUs offer the programmer a very powerful mechanism to offload computationally intensive portions of an algorithm." In addition to embedded HPC services, nCore Design offers on-site OpenACC and CUDA training courses and can help people determine which approach is optimal for their project.
Volcano: Eldad Klaiman
Eldad Klaiman is a signal/image processing engineer at Volcano Corporation, a medical device company focused on cardiovascular care. Eldad works on an R&D team that develops software to assist physicians who treat coronary disease in real time with devices such as wires, catheters and stents. Eldad and his team use GPUs and CUDA to accelerate image processing algorithms. "I like the simplicity of CUDA. It's maintainable and easy to understand," says Eldad. "CUDA helps me make the impossible a bit more possible."
ART: Monica Syal
Monica Syal is an aerospace engineer at Advanced Rotorcraft Technology (ART). She is working on development of a real-time rotorcraft brownout simulation for flight simulator applications, in collaboration with the University of Maryland (UMD) at College Park. The project is being funded by the Air Force Office of Scientific Research (AFOSR) under a Multidisciplinary University Research Initiative (MURI) grant. "The brownout problem is one of great significance in rotorcraft flight operations, and it is critical to explore ways to mitigate this very serious problem," says Monica. She adds: "Our ultimate goal in this research project is to achieve real-time speedups and, to this end, we will now be implementing most of our code on multiple GPUs by using OpenMP, MPI and CUDA."
Beckman Coulter: Bob Zigon
Bob Zigon of Beckman Coulter is responsible for investigating technologies that affect the flow cytometry, particle characterization, analytical ultracentrifugation (AUC) and automation business units. He says: "CUDA and Tesla are disruptive technologies. When they are applied to our problems we are capable of returning answers to clinicians and researchers in a fraction of a second. This causes people to change the way they interact with the data. Instead of looking at the data from 100,000 white blood cells, researchers can now manipulate five million cells." Bob is currently working on a prototype of a new CUDA-based application that will calculate the molar mass, gross shape and size distribution of protein samples by way of AUC.
Adobe: Steve Forde
Steve Forde is responsible for Adobe's visual effects product line, including Adobe After Effects (Ae) in Creative Suite 6, which offers a new GPU-accelerated 3D ray-traced compositing workflow capability. This enables motion graphics artists to quickly design realistic geometric text and shapes directly in a 3D space, eliminating the traditional, time-consuming need for external 3D tools. Steve says: "We created an environment in conjunction with NVIDIA that allows motion designers to quickly create high-fidelity motion graphics inside Ae -- without the computational expense -- by fully utilizing the GPU. This allows for much greater creativity while still being able to make the deadline."
Chaos Group: Vlado Koylazov
Vladimir "Vlado" Koylazov is co-founder and head of software development at Chaos Group, the makers of V-Ray and V-Ray RT. Chaos Group was founded in 1997 in Bulgaria. The company's interactive ray tracing software, V-Ray RT, leverages GPUs and CUDA to create photorealistic computer-generated imagery in real time. "Our software is used by artists to create so many amazing things. Whether it's a new electric vehicle, an energy efficient building or a blockbuster film, we're providing the tools to help artists visualize their imaginative designs," says Vlado.
Petrobras: Paulo Souza
Paulo Souza is the lead HPC (high-performance computing) developer of seismic imaging codes at Petrobras, the Brazilian multinational energy company. Paulo started working with GPUs in 2006, when he ported seismic applications to CUDA. Paulo comments: "We've invested in five GPU clusters, including the Grifo04 built with Tesla M2050 GPUs. Grifo04 is the fastest supercomputer in Latin America." At SC12 in November, Petrobras received the 2012 HPCwire Readers' Choice Award for Best Use of HPC in the Oil and Gas Industry.
Colorfront: Mark Jaszberenyi
Mark is the CEO of Colorfront, a digital post production facility which he co-founded in 2000 with his brother Aron. The Colorfront On-Set Dailies system was used by EFILM on the set of the new James Bond movie, Skyfall. Mark comments: "Colorfront On-Set Dailies relies exclusively on GPU image processing in CUDA. We can take advantage of multiple GPUs for even higher performance." NVIDIA GPUs drive Colorfront products on a range of platforms, from the Retina MacBook Pro through the Mac Pro, and up to the HP Z820 with multiple GPUs to process 4K RAW files at over 100 FPS.
MathWorks: Brian Fanous
Brian is a senior engineer in the signal processing and communications group at MathWorks, a leading developer of technical computing software. MathWorks is most well-known for the MATLAB and Simulink products. Brian is the primary developer of the GPU support for toolboxes in the signal processing area. He says: "We want to put the benefits of GPU computing into the hands of domain experts."
Virginia Tech: Anders Eklund
Anders Eklund is a postdoc at the Virginia Tech Carilion Research Institute. He conducts research in the field of medical image processing, especially fMRI (functional magnetic resonance imaging). Anders utilizes the CUDA programming model and and runs his calculations through MATLAB.
KTH: Erik Lindahl
Erik Lindahl of KTH Royal Institute of Technology and Stockholm University is a project leader for GROMACS, the popular open-source molecular dynamics program. The next release of GROMACS will include strong CUDA support. "CUDA is a long-term stable development that we can trust," comments Erik.
Triradiate Industries: GPU-Based Systems for Medical Imaging
Supratik Moulik is the founder of Triradiate Industries, a software development firm. "GPUs provide a way for complex visualization and analysis tasks to be performed quickly and with easily-attainable hardware," he says. "This translates into faster and more readily-available diagnostic tools which allow doctors to spend more time on patient care."
HZDR: Computational Radiation Physics
Michael Bussmann leads a Junior Research Group at HZDR in Dresden, Germany. His research focuses on computational radiation physics for applications in areas such as radiation therapy. "With GPU computing we have seen dramatic speed up of our code, which for the first time allows us to have live simulations of realistic scenarios," he says. His team's projects include PIConGPU, one of the world's fastest particle-in-cell codes.
University of Minnesota: GPUs for Green Urban Planning
Pete Willemsen, Associate Professor at the University of Minnesota Duluth, is using GPU computing to better understand complex interactions between urban areas and the environment. "Through the use of GPU-based simulation we hypothesize that there are urban structures and well-placed green infrastructure that can help to minimize energy use while also minimizing air pollution exposure. We hope our results can help guide future green urban planning projects," says Pete.
Boston University: GPUs for Scientific Discovery
Lorena Barba, assistant professor at Boston University, is a computational scientist and fluid dynamicist. A strong advocate of GPU computing, she teaches a computational fluid dynamics course that focuses on interactive collaboration rather than in-class lectures. She says: "GPUs are without a doubt a disruptive technology in the world of high-performance computing."
Princeton: GPU-Accelerated Swarm Behavior
Princeton's Iain Couzin is an expert in the study of collective animal behavior. His lab uses experimental systems - from ants and locust swarms to schooling fish and even human crowds - to explore the fundamental principles that underlie collective behavior. "GPU computing has utterly transformed the science we can do," he says.
INFN: Using GPUs to Better Understand the Universe
Denis Bastieri leads the Fermi Large Area Telescope (LAT) team, which observes and analyzes high-energy gamma rays from galaxies, black holes, pulsars and supernovae. Providing computational capabilities at one tenth the cost of conventional systems, CUDA GPUs allow him to accelerate his research and reduce the raw data coming from the satellite (around 120 GB/day) into meaningful physical information.
HP Labs: Big Data Analytics for Next-Gen Business Intelligence
According to HP Labs' Ren Wu, CUDA is a "game-changer" that enables the rapid analysis of massive volumes of business intelligence data. Using GPUs to accelerate big data analytics on multiple scales, HP Labs has achieved a 5-20x performance advantage over a pure CPU approach. GPUs will enable Ren Wu to gain new insights in the understanding of a number of critical areas, such as the environment, human health and global financial systems
Univ. of Pittsburgh: Searching for New Treatments
Postdoc Fellow Joshua Adelman is harnessing the computational power of CUDA GPUs to better understand and treat diseases, including ALS, epilepsy and Type 2 diabetes. Using molecular dynamics
, Adelman is simulating transport proteins that may hold the key to the development of new and more effective treatments. GPU acceleration has provided a several hundred-fold increase in protein simulation throughput.
Virginia Tech: Computing the Cure for Cancer
Virginia Tech is the inaugural research partner for the NVIDIA Foundation's Compute the Cure initiative. Team leader Wu-Chun Feng comments: "We plan to use the award to fundamentally change the way cancer biologists conduct their science....Supercomputing is no longer just the domain of big-iron supercomputers but also of personal desktop or deskside supercomputers. The domain area that I can foresee really benefiting the most from heterogeneous computing is the area of personalized medicine, which tailors healthcare to individual patients."
San Diego Supercomputer Center: Molecular Dynamics for Drug Discovery
Ross Walker developed AMBER
, a molecular dynamics (MD) software package for the simulation of biomolecules. Simulations help bio-physicists and computational chemists drive scientific discovery, such as creating more effective drugs to treat a range of diseases, e.g., the H1N1 virus. With GPU acceleration, AMBER is helping researchers dramatically speed up the process of developing better treatments.
University of Otago: Using Photonics to Detect Cancer
Biophotonics leverages optics/light to enhance research in medicine and the life sciences. Today, histological analysis with microscopy is a primary methodology for cancer diagnosis. However, it can be difficult to identify the type of cancer. CUDA GPUs enable Alexander Doronin to accelerate simulations by 1000x and to accelerate new biophotonics techniques.
Linkoping University: GPU-Accelerated Medical Image Processing
Anders Eklund specializes in medical image processing
. His area of interest is functional magnetic resonance imaging (fMRI), which identifies brain activity from magnetic resonance images as a means to identify and treat a variety of brain afflictions. Eklund has deployed GPU computing to save five years of processing time during the development and testing of a new algorithm for non-parametric statistical analysis of fMRI data.
Elemental Technologies: Real-Time Video for TVs, PCs, Mobile Devices
Jesse Rosenzweig and his team at Elemental Technologies have developed high-performance video algorithms for heterogeneous GPU/CPU architectures. Elemental integrates CUDA GPUs to decompress, process and recompress content in the video processing
pipeline. The resulting system allows media companies to deliver high-quality video streams of live events, satellite feeds, sports and more to TVs, PCs, tablets, and other mobile devices in real-time.
DTU: Designing Energy Platforms on the Oceans
Allan P. Engsig-Karup and researchers at the Technical University of Denmark (DTU) are focused on estimating the flow kinematics
and design loads on ocean structures, such as ships, oil platforms, energy devices, where predictions are required for the maximum expected lifetime load. The team has achieved impressive scalability results for the GPU-accelerated implementation of its OceanWave3D wave simulation model.
Mosaic ATM: Cost-effective Dynamic Airspace Configuration
Bart Gallet is researching improvements for the efficiency and safety of air transportation. The design of airspace and traffic flows is based on manual processes, as custom computer systems typically required to process the vast amounts of data are too expensive. However, CUDA and GPUs provide a cost-effective solution, speeding up key airspace and traffic flow computations 14 times faster.
Optimal Synthesis: Using GPUs to Improve Air Traffic Efficiency, Safety
Dr. Monish Tandale is leveraging the power of GPUs to improve the efficiency and safety of the U.S. air traffic system. He says: "With increasing computational capability available at our disposal, we can develop and analyze highly complex traffic flow models and run optimization algorithms to generate optimal schedules and routing of aircraft. The goal is to use automation to allow more people to fly without being inconvenienced by congestion and delayed flights."
University of Massachusetts: GPUs Advance Search for Greener Energy
Dr. Mehdi Raessi's research is focused on multi-phase flows and free-surface flows with phase change, which aims to identify more-efficient, environmentally friendly energy devices. GPU-based systems enable Dr. Raessi and his team to study much larger problems at a level of detail that was not feasible before.
Stone Ridge Technology: GPU-Accelerated Science and Computing
Dr. Vincent Natoli develops and ports high performance technical applications for some of the biggest corporations in the world, primarily oil and gas
companies. With GPUs, Dr. Natoli and his customers are seeing application acceleration of up to 20x compared with CPU-only systems, as well as cost savings due to the reduction of infrastructure footprint and power consumption.
Temple University: GPU-Accelerated Molecular Dynamics
Dr. Axel Kohlmeyer focuses on molecular dynamics (MD) and visualization research. GPUs and CUDA have enabled the processing of large, complex research calculations that was impossible just a short time ago, spawning a new wave of creativity in the study of compound systems, such as large bio-molecules embedded into realistic environments such as membranes.
EM Photonics: CFD Modeling for the U.S. Navy
John Humphrey is a pioneer in GPU computing and inventor of the CULA GPU-accelerated linear algebra library. EM Photonics' customers include the U.S. Navy, which relies on sophisticated CFD (computational fluid dynamics
) modeling to simulate landings and takeoffs based on variables such as light conditions, weather and approach angles. GPUs help decrease simulation time by an order of magnitude as compared to conventional systems.
Johns Hopkins University: Tools for Microsurgeons
While Kang Zhang was a PhD candidate at Johns Hopkins University, he used CUDA GPUs for research in interventional Optical Coherence Tomography (OCT) for microsurgery. OCT is an imaging modality capable of non-invasive 3D micrometer-resolution imaging, which makes it highly suitable for guiding microsurgery. Kang developed an ultra-high-speed, real-time imaging system using a hardware-software platform based on GPU technology.
UCSD: GPU-Accelerated Electromagnetic and Micromagnetic Simulators
Vitaliy Lomakin and his team at the University of California, San Diego (UCSD) develop simulators for computational electromagnetics and micromagnetics that are essential in the design of components such as solar cells, radar and antennas and hard drives. GPU-based simulators have a significant predictive power that allows analysis, design, and validation of systems before and during fabrication of products.
Microway: GPU Computing Momentum
Veteran technology inventor Stephen Fried is the founder of Microway. Microway's solutions include WhisperStation-PSC and fully-integrated Tesla GPU-based clusters, which are used by scientists for a variety of applications, from designing aircraft and space vehicles to mapping the ocean floor. BioStack-LS, a CUDA/Tesla-based Microway product, is pre-configured for life sciences software, including AMBER and MATLAB
SINTEF: Modeling the World in Real Time with GPUs
André R. Brodtkorb is a research scientist at SINTEF in Norway, where he works on GPU acceleration and algorithm design for shallow water simulations, tsunami warnings and simulation of storm surges and dam breaks. GPU-based systems allow SINTEF to run simulations at higher resolutions.
University of Plymouth: Developing Robots with CUDA
PhD student Martin Peniak is creating robots that can develop cognitive capabilities through interaction with their environments. Martin is using a CUDA-enabled software called Aquila to develop complex artificial neural networks for the real-time control of the robot as part of iTalk (Integration of Action and Language in Humanoid Robots), which is his PhD thesis project.
Portland Group: CUDA C Compiler
Doug Miles and his team at PGI create software tools to maximize performance and portability of applications across Linux, Windows and OSX. The partnership of PGI and NVIDIA has offered flexible tools to developers, including CUDA Fortran
and the PGI CUDA C compiler
. Doug comments: Today's application developers need flexibility. They want to be able to create innovative apps that leverage parallel computing and then deploy these apps on a wide range of target systems."
ERM and Digisens: Reconstructing our World of 2.1 Billion Years Ago
Arnaud Mazurier of ERM and Francois Curnier of Digisens utilized CUDA technology for data reconstruction of newly-discovered ancient organisms in Gabon, Africa. A team of scientists used GPU-based computed tomography to reconstruct data more quickly (6-10 minutes with GPUs versus 12-15 hours on CPUs).
Nexiwave: The Future of Voice Search Accelerated by CUDA
Ben Jiang, CEO of Nexiwave, is using the power of GPUs to improve speech indexing for easy extraction of archived content. "Ninety percent of human communication is through speech," he explains. "The amount of spoken words that could potentially be indexed and searched is staggering." Speech indexing can be efficiently processed in parallel which means the GPU is a perfect fit for it.