Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Stefan Wild Project Name: UNEDF_Optimization Division: MCS Project title: Optimization for NUCLEI and other applications Associated funding: DOE ASCR/NP/NNSA SciDAC-4 project NUCLEI DOE ASCR SciDAC-4 Institutes FASTMath and RAPIDS DOE ASCR Applied Math ROMPR Other Systems: NERSC Science: The NUCLEI (NUclear Computational Low-Energy Initiative) project will build upon recent successes in large-scale computations of atomic nuclei to provide results critical to nuclear science and nuclear astrophysics, and to nuclear applications in energy and national security. The large-scale computations we envision will transform the fields of low-energy nuclear physics and astrophysics. Physics topics to be addressed include nuclear interactions and their uncertainties, ab-initio studies of light nuclei and their reactions and of nucleonic matter and its astrophysical properties. We will also fundamentally advance the studies of neutron-rich nuclei and the fission of heavy nuclei, and the key nuclear physics issues in neutron stars and tests of fundamental symmetries. Project description: We are building on our optimization successes under the SciDAC-2 UNEDF (Universal Nuclear Energy Density Functional) and SciDAC-3 project NUCLEI. In this project, optimizations over parameter spaces corresponding to energy density functional parameters will be performed using the Argonne-developed POUNDERS optimization algorithm. We will work with DFT codes HFBTHO and HFODD and ab initio codes such as MFDn. For our parameter optimizations, we typically have 10-20 parameters and require on the order of 300 simulations. Based on the current simulation time, we would thus require 120 8-core ranks for up to 6 hours, or 6k core-hours for each version of the functional we want to optimize. We expect the number of core-hours will remain the same when using the new (and faster) nodes on Bebop because additional calculations are being added to the HFBTHO simulations required. Past experience has let us to expect to do roughly 15 such optimizations in FY18, for a total of 90k core-hours for this task. We are increasingly implementing software best practices in our work and this includes developing regression tests for the DFT components of NUCLEI. We expect that our development, optimization for the KNL nodes, and regression testing will require 120k core-hours. We will also be performing sensitivity analysis and exploration of the design space, we continue to work to improve our load balancing for the efficient computation of simultaneous parameter sets on thousands of cores. We expect to use 120k core-hours for this task. We will also be running 3D imaging studies of our optimization algorithms. Although we have not tried the KNL nodes, results on Bebop show that we will require up to 1600 cores for 12 hours for the largest of these runs (and anticipate doing 5 such runs). We require 130k core-hours total for this task. POUNDERS is available in the Toolkit of Advanced Optimization (TAO), but we still prototype specialized versions of the algorithm using MATLAB/Octave on a single, master node. We will test updates to the TAO version on Blues and compare the performance and correctness against the MATLAB/Octave versions. As in the past, we will have 4-8 users for this project. Industry partnership: Project URL: http://computingnuclei.org Current FY Hours Used: undetermined amount New FY Requested allocation: 460000 Q1: 115000 Q2: 115000 Q3: 115000 Q4: 115000 Justification: Storage requirements: Thank You, The LCRC Accounts System