Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Anh Ngo Project Name: IntensitySim Division: MSD Project title: Simulation Support for APS Data with ab initio Hybrid Reverse Monte Carlo Associated funding: LDRD Other Systems: Science: Our high-level goal is to improve the methods for determining large-scale structure for materials imaged at the APS by using Hybrid Reverse Monte Carlo, a combination of DFT, energy, and X-ray intensity calculations. Project description: Development history: The Calculate_I software came out of the Discovery Engines LDRD for combining experimental science with simulation. The program is written in Fortran with MPI for parallelism. OpenMP integration is planned. The code is compute-intensive: it reads a small amount of data produced by experiment and writes out a ~1 MB atomic structure dataset at the end. The computational method used models and improves a simulated atomic struction. A Monte Carlo iteration process then proposes improvements. The potential improvements are evaluated based on an energy calculation and an X-ray scattering intensity comparison to experimental data. The programming model used is Fortran + MPI. The code has been tested on LCRC scales and is expected to scale beyond that. We think that the analysis of data recently produced at CHESS by Ray Osborn will produce a Nature Materials paper. This project will use our software, Calculate_I, supported by 2000 VASP calculations. The plan to run VASP 2000 times and Calculate_I once for each experimental data set. The project initially contains the developers: Anh Ngo and Justin Wozniak. We will perform structure optimizations based on APS experiments of mullite and NaXV2O5 with different magnetic configurations will be carried out using (2 x 1 x 2) bulk unit cells. By comparing different magnetic ordering energies we will conclude which configuration corresponding to the ground state energy. We will determine the ground state structure as a result of the density functional theory (DFT) calculations using the VASP package. Periodic boundary condition (PBC) will be imposed in order to maintain periodicity of the solid support. Gradient corrected PBE density functional will be used for these calculations. From there, we can run Calculate_I to optimize the large scale structure. Those programs are currently available on Blues cluster and scale very well. Our total request is 1,000,000 core hours divided over the 4 quarters. Recent research products include: "Mechanical Flexure and Vibrational Maps of Individual Sexiphenyl Molecules on Ag(111)" Anh T. Ngo, Raymond Osborn, Justin Wozniak, and Peter Zapol Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 970000 Q1: 250000 Q2: 250000 Q3: 250000 Q4: 220000 Justification: We plan to use two codes: VASP and Calculate_I. The current version of Calculate_I scales well to 200 cores, which is the level at which we plan to run it. Storage requirements: 100GB Thank You, The LCRC Accounts System