LLNL logo Lawrence Livermore National Laboratory

Co-design at Lawrence Livermore National Lab

LLNL ASC Co-design Activities

The LLNL ASC Co-design Project is an umbrella for a number of efforts, all of which are designed to bridge the application teams to the research and vendor communities planning next generation extreme scale computing. Developers on the Engineering and Physics Integrated Code (EPIC) teams are under continuous pressure to improve the predictability and performance of their applications, as the EPIC codes are in continuous mission delivery. The Co-design project is a way to ensure that those teams have a seat at the table as new architectures, programming models, and other research are proposed. Likewise, their influence on the development of those technologies is critical, so they are usable to the developers responsible for keeping these codes running on the most sophisticated and complex supercomputing architectures.

A typical EPIC code can be over a million lines of code, with reliance on ten or twenty 3rd party libraries for performing tasks like linear solvers, domain decomposition, I/O, performance analysis, material databases, etc... Practical solutions that emphasize performance portability, robustness, and longevity are a priority for ASC co-design. At the same time, understanding long term trends in research and vendor roadmaps are a critical element of co-design that informs the direction these codes must pursue. Co-design is a two-way street designed to inform both the application developer and the system designers.

Sample Projects

The Advanced Architectures Software Development project is an effort that encourages EPIC code team members to break away from their day-to-day job to explore high-risk, high-reward ideas in the context of large scale code development. This includes exploration of new programming models, algorithms, or any element of research perhaps still in the pipeline that needs to be considered for an additional push into technology readiness that EPIC code development often demands.

AATEMPS (Advanced Algorithm Technology for Exascale Multiphysics Simulation) is an LDRD between the WCI (Weapons and Complex Integration) and Computation directorates at LLNL. It is exploring and developing the metrics, expertise, and insight required to transition LLNL's flagship EPIC applications to the next generation of advanced computing architectures. AATEMPS has been actively developing proxy applications to help streamline the process of analysis, with an ultimate goal of feeding those lessons learned back into the EPIC code teams.

The Scalable Applications Preparation project was developed to prepare application code developers to run on the 20Pf Sequoia and 5Pf Vulcan systems at LLNL. These BlueGene/Q systems are a dramatic leap in the number of cores available in any single system, and have demonstrated that MPI codes can scale to 1.5M processors. At the same time, the low-power A2 cores are demonstrating the need for fine-grained threading and SIMD vectorization. Both of these attributes are important considerations for future architectures, and Sequoia preparation is also next generation platform preparation.

Contact Us

Are you a vendor or researcher interested in exploring co-design with LLNL?

For additional information, please contact our team.