Time granted, just need to create the file set this week still. -- John Roberts Argonne National Laboratory CELS Systems [email protected] On 3/21/17, 12:40 PM, "[email protected] on behalf of Bair, Raymond A." <[email protected] on behalf of [email protected]> wrote: What we could do here is to scale the 250K request (99%?), and enable use of the time between now and 6/30. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 3/20/17, 4:24 PM, "sjk" <[email protected]> wrote: You are correct. I didn’t realize that Q2 ends in 10 days. Lets just shift it to 250 K in Q3 and 250 K in Q4 and nothing in Q2. Thank you for pointing this out. On Mar 20, 2017, at 4:10 PM, Bair, Raymond A. <[email protected]> wrote: > Stephen, > > I want to make sure I understand your request. > > Q2 is Jan-Mar, Q3 is Apr-Jun, etc. > > You requested 99K to use in the next 10 days. Is that what is intended? > > Regards, > > Ray > > ----------------------------------------- > Ray Bair > Argonne National Laboratory > and the University of Chicago > > > On 3/20/17, 1:39 PM, "[email protected] on behalf of [email protected]" <[email protected] on behalf of [email protected]> wrote: > > 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: Stephen Klippenstein > Applicant's institution: ANL > Applicant's division: CSE > Project Name: PACC > Project title: Predictive Automated Combustion Chemistry > Associated funding: DOE-ASCR > Other Systems: group owned linus cluster (48 nodes) > group owned nodes on blues (30 nodes) > I have access to about 1/5 of the CPU on these nodes > Science: To automatically generate high fidelity combustion mechanisms very high-level ab initio theoretical chemical kinetics. > Project description: The recent, largely DOE-supported, rise of theoretical gas phase chemical kinetics to a predictive science allows us to move beyond the empiricism (and concomitant limitations) of all prior mechanism development efforts. We are developing a fuel chemistry code that automatically generates high fidelity combustion mechanisms for user-selected fuels via large-scale, high-level, a priori theoretical chemical kinetics predictions. This code will exploit massively parallel computer architectures to simultaneously examine the hundreds or thousands of relevant chemical reactions, ushering in the wholistic development of mechanisms. Currently, mechanisms are generally validated against experimental data that often falls outside the range of end-use conditions and are developed sequentially one reaction at a time. The enhanced fidelity of a high level theory-derived mechanism, coupled with uncertainty driven improvement schemes, can directly access arbitrary > conditions with dramatically improved and rapidly constructed mechanisms for the end-use conditions. > > The procedure we are developing will involve the iterative application of mechanism generator routines (Reaction Mechanism Generator, RMG), an a priori theoretical kinetics routine (EStokTP), mechanism reduction routines (DRGASA) and chemical system solvers, and uncertainty quantification/global sensitivity analysis (UQ/GSA) software (Dakota). The initial step involves the generation of a reference set of reactions for the mechanism through an automated mechanism generator code (RMG). Currently in RMG, many, if not most, of the rates in this reference mechanism would be generated from crude rules based on analogies from limited data. We will replace these rough estimates with a priori theoretical kinetics predictions (EStokTP) employing a modest level of theoretical kinetics readily implemented for large numbers (1000s) of reactions. The EStokTP code will call various electronic structure codes such as GAUSSIAN, MOLPRO, and/or NWCHEM. With improved rates, RMG will be rerun to > generate a more reliable mechanism and associated theory-based estimates of the uncertainties in the predicted kinetics. The next iterations will selectively refine the mechanism by using GSA techniques under Dakota direction to shrink the uncertainties in mechanistic predictions (through further higher level theoretical evaluations) of key quantities (e.g., ignition delay, flame speeds, and speciation profiles) under arbitrary conditions. The Swift workflow software will be used to control all these codes. > > We with to use blues as the testbed for the initial stages of development of this code. We do not have information on the performance of the code as it is still being developed. However, most of the CPU will be spent on electronic structure calculations that use codes which are commonly used on blues already. Some parts of the code will be single core calculations, but that will likely be only on the order of a few% of the effort. There will be on the order of 10-20 project members. Mostly we will be running jobs that use ~10 to 20 nodes for a few hours at a time. > > > > Industry partnership: > Project URL: > Requested allocation: 499000 > Q1: 0 > Q2: 99000 > Q3: 200000 > Q4: 200000 > Justification: > Storage requirements: > > > The requester has used undetermined amount 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 > _______________________________________________ > allocations-admins mailing list > [email protected] > https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins > > _______________________________________________ allocations-admins mailing list [email protected] https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins