Comprehensive Simulation Lifecycle Management for High Performance Computing Modeling and Simulation

There are significant logistical barriers to entry-level high performance computing (HPC) modeling and simulation (M&S) users. Performing large-scale, massively parallel computations on modern supercomputing platforms is a very challenging task in and of itself. It requires huge amounts of analyst resources to construct datasets, transfer files, build codes, submit and monitor jobs, and analyze and archive results. This workflow requires significant attention to detail, and it is easy to miss steps, set up incorrect file or directory structures for datasets, spend inordinate effort to understand each computing platform and its unique job submission requirements, and decide how and where to archive potentially terabytes of simulation results. Collaboration among engineers is hampered by non-centralized storage of results, permission issues, and the sheer size of simulation result sets. Even for seasoned, veteran HPC users, the complexity of the overall HPC use process can be a barrier to daily use. We propose a system to streamline this workflow, provide audit trails for data used in simulations, code versions used, and storage locations of results, as well as integrating tools for file transfer, job submission, batch visualization, and other tasks that require engineer's time that is better spent considering the simulation itself. Further, the proposed SLMIR application (Simulation Lifecycle Management – IllinoisRocstar) sets up the infrastructure for collecting simulation data across an organization or organizations, so that engineers and scientists may discover information generated by others in their research areas, potentially saving time and money by leveraging previous work and avoiding duplication. The envisioned system targets users of HPC tools in the field, rather than developers writing those tools, since engineers and scientists in manufacturing and engineering industries are heavily invested in the results of using the tools.

This project will provide pathways to three products available to NASA -- software, engineering services, and turn-key systems. The software system architecture will be extremely extensible, based on XML and database system definitions. As such, it is anticipated that the system can be tailored to specific NASA customer's cluster-based simulation needs, as industry moves to depending on simulation technologies for competitive advantage, and purchase clusters of their own. This market will be tied to Engineering Services, and if Phase III funding can be identified to support a strong open-source support organization at IllinoisRocstar, then an open-source engineering services model can be considered.

The goal of the SLMIR project is to provide interfaces and services that make HPC a viable resource for non-HPC experts to use. In addition to software, it is anticipated that many organizations do not want to deal with the expense and in-house costs of developing and maintaining the hardware necessary for HPC either. We anticipate configuring our software, along with customer-requested software on commercial systems, and testing, and delivering the complete solution to commercial clients. Consulting services will be available based on the extensibility of the proposed system. IllinoisRocstar has the broad-based experience with a wide variety of supercomputing platforms to allow support of the proposed system on platforms located at industrial locations, NASA, DoD components, DOE, and other private companies. We anticipate specializing interfaces and processes for companies to use their own in-house codes with the system.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.