This looks like a reasonable request for 150K. Scale by the usual scale factor? P.S. Previously 100K was added in December. ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 2/1/17, 5:03 PM, "[email protected] on behalf of [email protected]" <[email protected] on behalf of [email protected]> wrote: Hello, A change in allocation has been requested: Requester: barlo (Lorenzo Bartolucci) Project: GAS_injection Title: High-Pressure Gaseous Fuel Injection Research – in-nozzle flow and jet evolution Description: This project focuses on investigating the in-nozzle flow for a gaseous inward opening injector and the consequently under-expanded jet evolution. This is aimed to provide a better comprehension of the local phenomenon and how those affect the typical behavior of the under-expanded gaseous jet (such as mach disk location and width, local thermo-fluid-dynamics properties and mixing with the surrounding mixture). At the current stage, the mesh influence on the quality of the numerical results has been deeply analyzed. An optimum mesh strategy and size have been found as a compromise between the numerical accuracy and the computational costs. The x-ray radiography data has been used as a reference for the comparison. Preliminary results have been obtained for the steady state simulation showing a good agreement with the experimental data. Currently an improvement in the agreement in terms of jet spread and mass distribution is under investigation. Future work would aim to represent also the opening and closing transient to provide a detailed instrument of analysis of the high injection phenomena occurring during these unsteady phases. Simulations will be performed with the commercial code CONVERGE and computational grids of the order of 15 millions of cells which will require the use of 64-144 cores depending on the specific case. The project has one member (barlo) working full-time on it, with 1-2 cases running regularly. With an average estimate of 96cores per run and 2 cases the expected amount of core-hrs needed for this project for each quarter is 96 cores x 24 hrs x 2 runs x 90 days = 414720. Considering some queue and post-processing time our request is for 1000000 hrs total, equally distributed for each quarter. Current: undetermined amount Justification: Recent scalability test (June 2016) performed on Blues provided following results: Simulation of gas jet from inward opening pintle. Mesh size = 15,000,000 cells 1 node (16 cores) – walltime = memory problems 2 nodes (32 cores) – walltime = 45min 15sec à 100 % 3 nodes (48 cores) – walltime = 33min 18sec à 90 % 6 nodes (96 cores) – walltime = 21min 50sec à 70 % Requested: 150000 A specific reason has been given: Since the high computational resources needed for the detailed simulation of the gaseous injection I would ask for an extention of the corehours dedicated to the project in order to finalize the paper carrying out the latest simulations with high computation grid resolution. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System