The Visualization Group's mission is to assist researchers in achieving their scientific knowledge discovery goals through visualization and analysis, while simultaneously advancing the state of the art through our own research. Their objective is to develop new capabilities in high performance visualization, analysis, and related data-intensive technologies. The development of these capabilities is driven by the needs of contemporary computational and experimental science projects central to the mission of the DOE Office of Science.
Our collaborators are diverse, ranging from theoretical astrophysicists to computational biologists. All have a theme in common: the need to understand complex systems through visual inspection of simulation results.
What is Scientific Visualization?
The goal of scientific visualization is to help scientists view and better understand their data. This data can come from experiments or numerical simulations. Often the size and complexity of the data makes it difficult to understand by direct inspection. Also, the data may be generated at several times during an experiment or simulation and understanding how the data varies with time may be difficult.
Scientific visualization can help with these difficulties by representing the data so that it may be viewed in its entirety. In the case of time varying data, animations can be created that show this variation in a natural way. Using virtual reality techniques, the data can be viewed and manipulated naturally in a true three dimensional environment (e.g. depth is explicitly perceived and not just implied).
All these techniques can allow scientists to better understand their data. Viewing the data in this way can quickly draw the scientist's attention to interesting and/or anomalous portions of the data. Because of this, we encourage scientists to use scientific visualization from the beginning of their experiments and simulations and not just when they think they have everything operating correctly. This also allows the scientists to develop a set of visualization tools and techniques that will help them understand their data as their research matures.
Facilities and Hardware
Over the years, the Visualization Group has installed a variety of equipment to enhance the researcher's ability to make discoveries using 3d visualization. Our pride and joy is the visualization lab, in 50B-1232 and 50B-1232A. This lab, which came on-line in May '96, contains our most sophisticated graphics hardware and display technology. The lab's 6'x8' rear projection screen and SGI Onyx workstation provide high resolution, high speed 3D graphical output for our visualization applications.
For the NERSC user community, we refer you to a resource called davinci.nersc.gov, which is a SGI Altix intended for interactive visual data analysis. For more information, please refer to the appropriate web page at the NERSC website.
Effective in mid-2000, we no longer provide software libraries or applications for the LBNL scientific community. A complete listing of visualization software that is available on NERSC platforms is listed on the appropriate NERSC web page.
Images/Movies on this site?
We are periodically asked about use of images and movies on this site. The short answer is that it is OK for you to use images/movies on this site for educational or non-commercial purposes, subject to a few simple terms.
The lab has established an End User License Agreement, which is at this url: http://www.lbl.gov/EndUserLicenseAgreement.html. Please read that Agreement carefully. It also explains what needs to happen if the intended use is commercial purposes; the lab has a procedure to follow in that case as well.
We do request that if you intend to use one of our images that you please inform us. And, when you use our images, please be sure to include the attribution line:
Image courtesy of the Lawrence Berkeley National Laboratory Visualization Group, and is ©University of California, Lawrence Berkeley National Laboratory.
The Visualization Group was created in 1990 (it was then called the Graphics Group) when Wes Bethel was hired to work under Nancy Johnston to tame, explore and apply to ongoing scientific programs some brand new software (AVS) that was running on some brand new hardware (the Ardent Titan, a multiprocessor graphics machine that is remarkably similar in architecture to the modern-day Onyx..). Prior to this point, all of the graphics work done at the lab was accomplished via making call from within code to subroutine libraries, or using graphics "packages" such as gnuplot. This step marked the lab's first foray into interactive 3d visualization that was supported by some serious graphics hardware. The Visualization Group's mission, circa 1990, was to team with LBNL researchers to accelerate scientific understanding using a new breed of visual data analysis hardware and software.
The initial success of the graphics program was due in part to the diligence of the researchers involved. They were a tough, tenacious bunch who really wanted to use state-of-the-art 3D visual data analysis, and were willing to put in some hard work to realize its benefits. Our experience, which is confirmed by the experiences of our peers in other visualization programs, has been that the most successful visualization projects involve a close working relationship between scientific researcher and computer (visualization) scientist. The jointly-developed visualizations helped push the envelope of some scientific programs. For example, using 3d visualization, researchers in Earth Science were able to verify the existance of a deep subduction zone near New Hebrides. One of our awards was given for a compelling visualization from a hydrological study used in modelling nuclear waste storage facilities in Sweden. The theme throughout all of our activities is the symbiotic and synergystic relationship between the visualization technology and the scientific research program.
As the visualization challenges began to grow, over time our work began to branch out to include more research and development. For example, we created the VR Library, which is a suite of AVS modules designed to allow users to easily incorporate VR input devices into the user interface of a visualization network. Another large project has been our input to the Khoros system, which is similar in many ways to AVS. However, as our interests and workload have increased, we have also expanded our staff, so our commitment to collaborating with users on visualization projects has not changed.
With the arrival of the NERSC Center to LBNL in 1995, the NERSC and LBNL Visualization efforts were combined. The focus of the NERSC visualization program is upon providing visual data analysis support to the NERSC user population, most of which is remotely located with respect to the NERSC facility. Such support comes in the form of a combination of resources. Some of the resources are graphics and visualization programming libraries, like the NCAR Graphics library. Other production visualization support comes in the form of interactive 2D and 3D visualization applications that are installed on many of the NERSC platforms. Still yet another form of support comes in the form of understanding the needs of a community of remote users with challenging visualization needs, and using those needs to guide fundamental visualization research activities.
Starting in about 2000, the LBNL Visualization Group ceased to receive funding from LBNL for the purposes of providing direct one-on-one support to the LBNL scientific community. This unfortunate turn of events reflects shrinking budgets within the lab community, among other factors. One of the difficulties we face, which is not unique to our program, is the challenge in quantifying the benefit of visualization to scientific research. From a financial management perspective, a valid question is "how much stuff do we get for each dollar of funding?" The challenge is in defining stuff, as well as the fact that like any research activity, there is an element of unpredictability in the program. It was taking us more time to justify and defend that budget than to actually help the lab's scientific community. While we are not funded to provide direct visualization support to the lab's scientific community, we welcome the challenge of visualization problems for which there "is no solution," regardless of whether they come from the LBNL scientific or NERSC user communities. Unfortunately, we are not able to provide visualization support "to the public" (which we'll define as those who are neither LBNL researchers or researchers with a NERSC account).
Concurrently, DOE's investment in our base research program began to grow with an initial emphasis upon techniques for visualization of Adaptive Mesh Refinement data, in particular, and multiresolution visualization research, in general. Looking forward to the future, we anticipate a good deal of research activity in the area of component based visualization tools for use in multi-institution, remote and distributed and parallel environments