We have developed a novel analytic capability for scientists and engineers to obtain insight from ongoing large-scale parallel unstructured mesh simulations running on thousands of processors. The breakthrough is made possible by a new approach that visualizes partial differential equation (PDE) solution data simultaneously while a parallel PDE solver executes. The solution field is pipelined directly to volume rendering, which is computed in parallel using the same processors that solve the PDE equations. Because our approach avoids the bottlenecks associated with transferring and storing large volumes of output data, it offers a promising approach to overcoming the challenges of visualization of petascale simulations. The submitted video demonstrates real-time on-the-fly monitoring, interpreting, and steering from a remote laptop computer of a 1024-processor simulation of the 1994 Northridge earthquake in Southern California.

Growing compute capacity coupled with advances in parallel filesystem performance and stability mean that HPC users will inevitably create and store larger datasets. If data residing on parallel filesystems is not efficiently offloaded to archival storage, disruptions in the compute cycle will occur. Hierarchical storage caches are a vital aspect of the HPC storage machine because they offload and prime the compute engine's parallel filesystem. To keep pace with expanding HPC data systems, these caches must adapt via new scalable architectures. The Pittsburgh Supercomputing Center (PSC) has developed a cooperative caching architecture to act as a distributed front-end cache to a hierarchical storage manager. SLASH, the scalable lightweight archival storage hierarchy, provides means for creating parallel caching systems on an otherwise "monolithic" archival storage system without requiring modifications the archival system software.