FastBit in Visual Analytics

Colliding high-energy particles is a fundamental tool for probing the structure of elementary particles and understanding the building blocks of our Universe. However, the accelerators are becoming bigger (many miles long) and more expensive to achieve the increasing energy needs, and this is a challenge. The Laser Wakefield Particle Accelerator (LWPA) is a new type of accelerator that is capable of accelerating particles to very high energy in a much shorter distance. To better understand these accelerators, simulations are conducted using software, such as the VORPAL framework. Such a simulation can produce a very large amount of data. For example, a run of VORPAL may output 100 timesteps of 200 GB each. A volume rendering of the particles in such a simulation is shown in figure 4.

O. Rubel et al. 2008 Figure 4. A volume rendering of the density of particles in a laser wakefield particle accelerator.

FastBit has been used to quickly find the particles that have undergone wakefield acceleration and track them throughout the simulation. In addition, FastBit can also efficiently compute histograms needed for the histogram-based parallel coordinate display used for visualizing and selecting particles. As in a normal parallel coordinate display, there is one axis (a vertical line in figure 5) for each variable in a dataset. However, unlike a normal parallel coordinate display that draws a line for each particle from the simulation, a set of two-dimensional histograms are used to provide the density of lines that would have been displayed. This reduces the amount of work needed to produce the parallel coordinate display and enables us to work with much larger datasets. FastBit can efficiently filter the particles to be shown in the parallel coordinate display. In figure 5 a focused beam is displayed (in red) on top of a gray background of a larger subset of particles. This filtering reduces the number of particles involved in the histogram computation and improves the overall analysis speed.

O. Rubel et al. 2008 Figure 5. A parallel coordinate display of a focused beam (red) and its background (gray) from a set of simulation data for a laser wakefield particle accelerator.

FastBit is also very efficient at tracking particles across time. In figure 6, the particles selected in figure 5 are shown at different timesteps of the simulation. In this task, FastBit indexes can dramatically decrease the execution time.

O. Rubel et al. 2008 Figure 6. Tracks made by selected particles in a laser wakefield particle accelerator. The gray dots indicate background particles in a timestep in the middle of the simulation.

Altogether, using FastBit has significantly increased the speed of this visual task. For example, in the first test run of the FastBit enabled tracking program, it was able to complete the tracking task in 0.3 seconds whereas the original tracking program was using 5 minutes, a 1,000-fold speedup in this case.