Ultrascale Visualization In Situ Visualization for Large- Scale Combustion Simulations Hongfeng Yu ■ Sandia National Laboratories

Supercomputers ’ massive processing power drives scientific discovery in many areas, and their computing power and storage capacity grow rapidly every year. As scientists gain access to more powerful machines, they attempt to solve larger, more complex problems. Consequently, the amount of data their scientific simulations generate is increasing at an astounding pace. Large-scale scientific simulations typically output time-varying multivariate volumetric data. Such data can now easily occupy petabytes of storage space. To maximize simulation output’s use, scientists need efficient, effective solutions to manage and study the ever-growing data volume. Supercomputer time is a prized commodity. Scientist