Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Shashi Aithal Applicant's institution: ANL Applicant's division: MCS Project Name: HSCD Project title: High Speed Combustion and Detonation simulations Associated funding: DoE/ASCR Other Systems: Ranger, BG/P (ALCF) Science: The primary goal of this project is to develop a fundamental understanding of the physical processes that bring about a transition from deflagration (slowly moving laminar flames) to detonation (fast moving turbulent flames) in combusting mixtures of fuels and oxygen. The key objective of this work is to conduct first-principles Direct Numerical Simulations for predicting Deflagration-to-Detonation transition (DDT) in reacting turbulent mixtures. Project description: DDT is inherently a multi-physics/multi-scale phenomena involving a close coupling of multiple physical and chemical processes, occurring over a wide range of spatial and temporal scales. Modeling and simulation of this phenomenon requires a detailed description of the various underlying physical and chemical processes such as chemical kinetics and energy release, heat conduction, species diffusion and interaction of turbulent shocks with boundary layers. A compressible reactive flow Navier-Stokes solver named High-Speed Combustion and Detonation (HSCD) code will be used for the simulation of the DDT phenomena. The HSCD code is a distributed memory parallel adaptive mesh refinement (AMR) reactive flow Navier-Stokes code. A distinct feature of the code is a dynamic cell-by-cell AMR based on a parallel fully threaded tree (FTT) structure. Architecturally, the code consists of three separate layers: (1) the FTT library which provides general services related to all parallel aspects of the code’s execution; (2) the reactive Navier-Stokes AMR code, ALLA, which rides on top of FTT and is responsible for numerical integration and AMR; (3) problem-specific algorithms such as material properties routines (equation of state, kinetics, microscopic transport) and problem initialization routines called from ALLA. Computations in ALLA are organized as a set of global computational steps, with each step followed by communication work which synchronizes data across the domain decomposition boundaries. Application algorithms are programmed in terms of work-functions which are passed to and executed by the global parallel iterator. The code is parallellzed using a hybrid OpenMP/MPI strategy. On each MPI rank the iterator parses the mesh and passes small chunks of cells to a work-function until the entire mesh is processed. The loop over the cells in a chunk is performed inside the work-functions themselves. These loops are parallelized using OpenMP. HSCD scales well up to 131K cores on the BG/P. Based on our simulations on BG/P, it would take about 2 sec/time-step on 1K cores on Fusion and a complete run would require about a 100K time-steps. Thus each run would need about 50K core-hours. We are planning a series of 8-10 parametric cases to identify important operating conditions and flow regimes. The code needs MPI, OpenMP libraries, C and Fortran 90 compilers. It is expected that the project will have 4-5 members including the PI. Project URL: Requested allocation: 450000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System