January 2017 - Allocations-admins - mailman3.cels.anl.gov

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 31 Jan '17

31 Jan '17

Hello, A change in allocation has been requested: Requester: dshaver (Dillon Shaver) Project: CMFD-RANS Title: Development of Multiphase RANS turbulence models Description: RANS and uRANS turbulence models are necessary for the accurate CFD simulation of flows on large scales, such as nuclear reactor fuel rod bundles and full cores. The current state-of-the art for simulation of boiling flows utilizes the RANS framework. Recently, the k-omega turbulence model has been demonstrated with Nek using a homogeneous equilibrium model of multiphase flow (thanks significantly to LCRC support). The goals of this project will be to further extend that model to more sophisticated formulations of multiphase flow, specifically the two-fluid model available in Nek-2P. This will enable the further development and validation of multi-phase models with Nek and the subsequent simulation of large, complex problems of interest in both science and industry. Additionally support for the development of the k-omega model will include simulations of a twin-jet mixing facility to evluate the model performce in capturing the effects of buoyancy and free -shear layers on turbulence. Both of which can play a signficant role in multiphase flows. This project will use the CFD code Nek5000 and its branch, Nek2P. Written in a combination of Fortran and C, Nek has been demonstrated to be a highly scalable code using the MPI library. Nek5000 has a long history of use and development on the LCRC systems. Proposed simulations will include helical pipe heat exchangers, nuclear reactor subchannels, and the turbulent mixing of twin jets. Meshes for these problems range in size from <100K elements to >1M elements. The project is expected to have between 2-4 members. Current: undetermined amount Justification: Requested: 250000 A specific reason has been given: The required simulations became significantly larger than initially anticipated and the design space has increased. With the success of the two-fluid model in Nek-2P, it is now being used for some limited parametric testing on the helical pipe heat exchanger. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Project Request: KTP
by Bair, Raymond A. 31 Jan '17

31 Jan '17

New project KTP. The Q2 request is 86K, which we could approve. Q3 and Q4 are 200K each – can be left to the allocations committee. ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/30/17, 1:13 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov" <allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov> 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: Ahren Jasper Applicant's institution: ANL Applicant's division: CSE Project Name: KTP Project title: Temperature- and Pressure-Dependent Kinetics Associated funding: DOE BES Other Systems: Chemistry's Blues cores b431-b460 Chemistry's linux 1300-core cluster "Linus" Science: In recent work, we have shown that quantitative predictions of gas phase chemistry can be made using a combination of classical trajectories (molecular dynamics), transition state theory, and master equation calculations. The present project focuses on using classical trajectories to predict energy transfer parameters for use in master equation calculations for real fuels of relevance to combustion. Whereas in past work, we have considered small test case fuels, such as methane, real fuels often contain much larger species such as dodecane and the large esters that appear in biodiesels. The main goal of this project is to use high performance computing to adapt our methods to the treatment of larger (real) fuels. Project description: We will (1) fit full-dimensional potential surfaces, and (2) run trajectories to calculate energy transfer. Step (1) involves parallel sampling of the intermolecular potential (which has a dimensionality of 3, 5, or 6 for atomic, diatomic, and polyatomic baths, respectively) using commercial quantum chemistry (QC) programs (principally Molpro). "Good" QC methods, such as counterpoise corrected CCSD(T)/CBS, can be prohibitively expensive, even in large parallel computing environments. We will therefore continue our work optimizing (a) the choice of QC method (considering MP2, dispersion corrected DFT, etc) and (b) sampling point placement (considering biased distributions and quasirandom (Sobol) sampling). In preliminary work, we have found (a) that MP2 is often accurate for Ar as a bath gas, and (b) a biased quasirandom sampling can increase efficiency by a factor of 15. Still, for large molecules, sampling will be expensive, requiring ~10000 1-hour QC calculations. Fitting the resulting data will be done using a genetic algorithm or linear least squares fitting. The cost of the fi tting is negligible relative to the cost of sampling. Step (2) involves using our parallel trajectory code DiNT to run energy transfer trajectories. 10,000 independent trajectories are required to characterize energy transfer for each system of interest, with each trajectory taking 1-60 minutes, depending on the size of the system. In this first step, we will consider medium sized alcohols, peroxides, and esters. For each of these three class, we will consider 9 cases (e.g., the three alcohols ethanol, butanol, and hexanol for each of the three baths He, Ar, and N2) for a total of 27 cases. The cost of both steps 1 and 2 will scale dramatically (in different ways) based on system size, with the largest systems considered requiring most of the computer time. For example, we estimate 10000 samples of the hexanol+Ar potential to require 5000 hrs, though we likely have to "waste" processors due to I/O limitations, thus increasing the core-hour cost by up to a factor of 3. Altogether, the quantum chemistry costs should require: 3 classes * 9 cases/class * 10,000 core-hours/case = 270,000 core-hours. In the second step, for each of the 27 cases, we will run 10,000 trajectories. These take 30 minutes, on average, for a total of 135,000 core-hours. The trajectory calculations run at near 100% parallel efficiency and scaling. Steps 1 and 2 thus require a total of 405,000 core-hours, to which we add 20% for testing and development, leading to a total request of 486,000 core-hours. Industry partnership: None Project URL: Requested allocation: 486000 Q1: 0 Q2: 86000 Q3: 200000 Q4: 200000 Justification: Storage requirements: Less than 1 TB 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 allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] [LCRC Accounts] Project Allocation Request
by Bair, Raymond A. 31 Jan '17

31 Jan '17

This request is consistent with the project size. Recommend we add 15K to AWA-beam-dynamics ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/30/17, 5:47 AM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov" <allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov> wrote: Hello, A change in allocation has been requested: Requester: neveu (Nicole Neveu) Project: AWA-beam-dynamics Title: High Accuracy Beam Dynamic Simulations Description: The primary goal of this project is to simulate the two linear electron accelerators at the Argonne Wakefield Accelerator facility. To date, a full 3D models of the beam lines has not been simulated. Using OPAL-T and the parallel capabilities of Blues, we seek better agreement between simulations and experiment by accounting for 3D effects, space charge, coherent synchrotron radiation, and wakefields. Current: undetermined amount Justification: Requested: 15000 A specific reason has been given: The parameter space for one of our current projects is bigger than expected. A large random sample was done to characterize the space. This took the remainder of our hours. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] Project Allocation Request - Mullite
by Bair, Raymond A. 31 Jan '17

31 Jan '17

On the CSAC report, it does not appear that this allocation was made yet. Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/25/17, 8:47 PM, "Ngo, Anh Tuan" <ango(a)anl.gov> wrote: Hi Ray, As I mentioned, we plan to run 60 calculations more and hence, our additional request is 161280 CPU hrs. Could you please help us? Thanks and best regards, Anh ________________________________________ From: Bair, Raymond A. Sent: Wednesday, January 25, 2017 5:37 PM To: Ngo, Anh Tuan Cc: Roberts, John E. Subject: Re: Project Allocation Request - Mullite Dear Anh Ngo, We need to know how much additional time you need between today and March 31. In addition, there will be new allocations starting April 1, based on your FY2017 request last September. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory On 1/25/17, 4:40 PM, "Ngo, Anh Tuan" <ango(a)anl.gov> wrote: Hi John, Sorry, I might not understand your email. We have received allocations for mullite project in October, 1,2016 from LCRC but at the moment we run out time for this project that why I submitted request for time yesterday. Could you Please see email below? and let me know if you could help. Thanks and best regards, Anh ---------- Forwarded message ---------- From: <accounts(a)lcrc.anl.gov> Date: Sat, Oct 1, 2016 at 12:00 AM Subject: [LCRC Accounts] Your LCRC project has been given a new allocation To: ango(a)lcrc.anl.gov You have received this mail because you are the PI of the following project on the LCRC cluster: Project Name: Mullite Your project has been approved for an additional allocation: Allocation: 352,500 hours ========================================================================== Note from the LCRC Allocation Committee: This is your allocation for the first half of FY17 (Oct-Mar), granted on Oct 1. Due to high demand for computing time on Blues, we had to scale down the number of hours given to projects this FY. Half of this allocation (176,250 core-hours) must be used by the end of the first quarter of the FY (Dec 31). The Allocations Committee meets 4 times a year to review allocation requests. At the September and March meetings, they review all projects for a 6 month allocation. New project and additional allocation requests are reviewed at each meeting. Project balances are zero'd out at the start of the fiscal year. Allocations given for the first 6 months of the fiscal year become active on the first day of the fiscal year (October 1st). All other allocations given become active immediately. PIs can request more time for their project at any point during the year via the project management web pages. These requests, like the annual requests, are reviewed by the committee in the context of the overall usage of the cluster. ========================================================================== For more information on projects and using the allocation tools refer to the LCRC documentation: http://www.lcrc.anl.gov/info/ManagingAllocations Thank You, support(a)lcrc.anl.gov ________________________________________ From: Bair, Raymond A. Sent: Wednesday, January 25, 2017 3:26 PM To: Ngo, Anh Tuan Cc: Roberts, John E. Subject: re: Project Allocation Request - Mullite Dear Anh Ngo, This project has not time in FY2017 because it appears that no one turned in a request for time. We will use the information below for Q3 (April-Jun) and Q4 (Jul-Sep). Please let us know if this is not what you intended. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/23/17, 5:31 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov" <allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov> wrote: Hello, A change in allocation has been requested: Requester: ango (Anh Ngo) Project: Mullite Title: The structure of disordered Mullite studied using a ab initio hybrid reversed Monte Carlo algorithm Description: Understanding the nature of the disordered structures and physical/chemical properties of mullite is therefore an area of increasing importance and has been the focus of investigation for many decades. In this project we would like to give a complete description of the real mullite. We will use ab initio energetic cluster expansion techniques to explore all distinct interactions/mechanisms between/of Al, Si, Oxygen and vacancy in the mullite. We then integrate these effective interaction parameters to build a fully microscopic model with our new developed ab initio hybrid reversed Monte Carlo. The project will be realized by collaboration between Anh Ngo (MSD, argonne) and Peter Zapol (MSD, argonne) and Ray Osborn (MSD, Argonne). Our results obtained from ab initio hybrid reversed Monte Carlo can be directly compared to diffuse scattering data measured at APS from Osbornâ€™s group. To obtain our goal the following calculations will be performed: (a) The initial calculations will focus on finding the ab initio cluster expansion energies (CE) for all pair, triplet and higher order interactions. (b) To find real disordered structures, in our further investigations we will use CE obtained from step 1 to perform ab initio hybrid reversed Monte Carlo. This work will plan to use the high-performance computing facilities at LCRC. The first-principles calculations will be carried out using the electronic-structure code VASP. The Blues cluster is particularly suitable for this research. In order to obtain good Cluster expansion nergies we will plan to do 200 energy optimized calculations each calculation contains 60 atoms takes approximately 3072 CPU hours (16*4*48 ). The ab initio hybrid reversed Monte Carlo calculation takes roughly 26880 CPU hours (16*10*168) we plan to run 8 of these calculations. So our total request is 200*3072 + 8*26880 = 829440. In summary, the total requested allocations for this project is 829440 CPU hours. Current: undetermined amount Justification: VASP performance on Blues is well established, we use the latest version maintained on Blues. Requested: 161280 A specific reason has been given: Dear LCRC supports, We recently submitted our paper in physical review B : *Anh T. Ngo, Raymond Osborn, Stephan Rosenkranz, Justin Wozniak, and Peter Zapol: The Incommensurate Modulation in Mullite: An ab initio Approach* In order to reply to referee reports, we plan to run 60 more calculations, each calculation will approximately take 2688 hrs. Hence, our additional request is 2688*60 = 161280 CPU hrs. Best regards, Anh Ngo This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Project Allocation Request
by Bair, Raymond A. 31 Jan '17

31 Jan '17

This is a request for Q3 and Q4 ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/21/17, 1:42 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov" <allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov> wrote: Hello, A change in allocation has been requested: Requester: linyun (Linyun Liang) This is a request for the Project: Meso_Nuclear Title: Gas bubble kinetics in an irradiated U-Mo using a mesoscale simulation approach Description: A mesoscale simulation approach is developed to study the gas bubble evolution kinetics in irradiated U-Mo alloy fuels. Density functional theory is first utilized to predict the defect formation energies, the Xe-U-Mo interfacial energy, and the elastic constants of U-Mo alloys, among other quantities. These obtained material properties are then incorporated into a mesoscale model to study the kinetics of gas bubbles in U-Mo under irradiation. By developing a new free energy model of Xe gas bubble based on the van der Walls equation of state, the gas bubble pressure can be estimated. The effects of vacancy and self-interstitial atom (SIA) diffusion on the gas bubble pressure will be further investigated and validated. The events employed in the simulation are the absorption and emission of a vacancy, a SIA, and an interstitial U by a single Xe bubble, while an event associated with SIA loops or substitutional U is neglected here for simplicity. We will systemat ically investigate the effect of Xe, vacancy, and SIA concentration, fission defect generation and their recombination, sink strength, grain boundaries, and elastic interaction on the growth kinetics of gas bubble. The swelling of U-Mo due to fission gas bubble will be simulated and compared to experimental observations. Our study aims to provide insight into understanding of the structural damage in irradiated U-Mo materials. In our phase field model, we have four highly non-linear, coupled partial differential equations which must be solved numerically. In the meaning time, the elastic equilibrium equation is also included in each time step, which requires to obtain the elastic solution at the equilibrium state for each time step and can significantly increase the cost of the simulation time. And also in order to avoid the partial loss of interfacial energy, we will extend our model to three dimensional system. Therefore, to effectively solve this problem, we will need about 150,000 core hours including the 6,000 core-hour to debug and test the algorithm and code and 144,000 core-hours (4 cores/node * 4 nodes * 10 hours of run time * 800 different runs) for the nucleation and growth of gas bubbles calculations. We will use our developed FFTW code to solve the problem. By using the Fusion and Blues computing resources will greatly accelerated the code development and calculations. In conclusion, the total core-hours requested for this project throughout the FY2016 will be 150,000 core-hours. Fig. 1 Simulated results of gas bubbles nucleation and growth under the irradiated U-7Mo. Current: undetermined amount Justification: Requested: 150000 A specific reason has been given: I have zero allocation hours balance. I am requesting additional 50,000 core-hours for the third quarter and 100,000 core-hours for the four quarter. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] [LCRC Accounts] Project Allocation Request
by Bair, Raymond A. 31 Jan '17

31 Jan '17

Has this change been made? ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 12/8/16, 1:27 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Bair, Raymond A." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of rbair(a)anl.gov> wrote: Recommend that we approve 150K. ----------------------------------------- Ray Bair Argonne National Laboratory On 12/6/16, 10:51 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov" <allocations-admins-bounces(a)lcrc.anl.gov on behalf of accounts(a)lcrc.anl.gov> wrote: Hello, A change in allocation has been requested: Requester: ssom (Sibendu Som) Project: gm_gasexchange Title: Modeling Cyclic Variability in the GM TCC Engine Using LES Description: The project is divided into three parts: In the first part, large eddy simulation (LES) will be used to model 50 consecutive cycles of the fired GM transparent combustion chamber (TCC) engine with a base grid size of 0.5 mm. Adaptive mesh refinement will be employed to resolve the flame region with a 0.125 mm grid. The mean and variance of the predicted flow field and the in-cylinder pressures will be compared to the PIV measurements. In the second part of this study, the parallel cycle methodology will be extended to the fired TCC engine, which will enable fast throughput computation of the cyclic variability. In the previous year, this methodology was demonstrated for motored conditions. In the final part of this study, the parallel cycle methodology will be used to perform a parametric study on the influence of numerical parameters like minimum grid size, number of cycles, and spark energy deposition on the predicted cyclic variability. Current: undetermined amount Justification: Requested: 150000 A specific reason has been given: we ran more simulations than what we were expecting to run initially. These extra simulations have resulted in an additional paper with GM R&D for SAE WC 2017. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] User Time Request
by Bair, Raymond A. 31 Jan '17

31 Jan '17

Has this change been made? ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 12/8/16, 8:37 AM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Bair, Raymond A." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of rbair(a)anl.gov> wrote: OK, lets do what he requests. ----------------------------------------- Ray Bair Argonne National Laboratory On 12/7/16, 11:12 AM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Roberts, John E." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of jeroberts(a)anl.gov> wrote: User never got back to me, so I just pinged him again and he gave back a prompt response this time: “Can I move the time from the second half of the year? Is it possible to move something like 100000 hours?” -- John Roberts Argonne National Laboratory CELS Systems jeroberts(a)anl.gov On 12/7/16, 10:43 AM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Bair, Raymond A." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of rbair(a)anl.gov> wrote: What did they decide to do? ----------------------------------------- Ray Bair Argonne National Laboratory On 11/30/16, 2:28 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Bair, Raymond A." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of rbair(a)anl.gov> wrote: We could offer to advance time from the 2nd half. That is an established procedure. Alternately we could add time to his project, though we would like hi to submit a request through the project system. Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 11/30/16, 12:41 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Roberts, John E." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of jeroberts(a)anl.gov> wrote: To clarify, he wrote quarter, but upon checking he’s almost out of time for the half year (Q1 and Q2). Total hours used: 202644.81 Current balance: 1355.19 -- John Roberts Argonne National Laboratory CELS Systems jeroberts(a)anl.gov On 11/30/16, 9:39 AM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Roberts, John E." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of jeroberts(a)anl.gov> wrote: Just following up on the below. Thanks! -- John Roberts Argonne National Laboratory CELS Systems jeroberts(a)anl.gov On 11/28/16, 4:55 PM, "allocations-admins-bounces(a)lcrc.anl.gov on behalf of Roberts, John E." <allocations-admins-bounces(a)lcrc.anl.gov on behalf of jeroberts(a)anl.gov> wrote: Received this request from a user today: “Our allocation time of dmft_for_oxides project in this quarter is almost depleted. Would it be possible to transfer allocation time for the next quarter to this quarter since we have more heavy calculations scheduled this quarter than the next quarter? Thank you for your understanding. Best Regards -- Hyowon Park Assistant Professor, Department of Physics, Materials Science Division, University of Illinois at Chicago. Argonne National Lab.” Thanks. -- John Roberts Argonne National Laboratory CELS Systems jeroberts(a)anl.gov _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] Fwd: [LCRC Accounts] Project Request: bldg_energy_modeling
by Bair, Raymond A. 31 Jan '17

31 Jan '17

Has this allocation been made? ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago From: <allocations-admins-bounces(a)lcrc.anl.gov> on behalf of Raymond Bair <rbair(a)anl.gov> Reply-To: LCRC Allocations Admins <allocations-admins(a)lcrc.anl.gov> Date: Tuesday, January 24, 2017 at 1:54 PM To: "allocations-admins(a)lcrc.anl.gov" <allocations-admins(a)lcrc.anl.gov> Subject: [allocations-admins] Fwd: [LCRC Accounts] Project Request: bldg_energy_modeling Sounds good to me. Ray Sent from my iPhone Begin forwarded message: From: "Bair, Raymond A." <rbair(a)anl.gov<mailto:[email protected]>> Date: January 16, 2017 at 11:56:34 AM CST To: LCRC Allocations Admins <allocations-admins(a)lcrc.anl.gov<mailto:[email protected]>> Subject: Re: [allocations-admins] [LCRC Accounts] Project Request: bldg_energy_modeling Reply-To: LCRC Allocations Admins <allocations-admins(a)lcrc.anl.gov<mailto:[email protected]>> I recommend that we allocate 10K for this project, where 10K is the minimum for future project allocations. Ray ----------------------------------------- Ray Bair Argonne National Laboratory On 1/14/17, 8:48 PM, "allocations-admins-bounces(a)lcrc.anl.gov<mailto:[email protected]> on behalf of accounts(a)lcrc.anl.gov<mailto:[email protected]>" <allocations-admins-bounces(a)lcrc.anl.gov<mailto:[email protected]> on behalf of accounts(a)lcrc.anl.gov<mailto:[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: Ralph T. Muehleisen Applicant's institution: ANL Applicant's division: ES Project Name: bldg_energy_modeling Project title: Building Energy Modeling Associated funding: DOE Building Technologies Office Other Systems: Science: Large scale energy modeling of buildings. Collections of hundreds to thousands of buildings of various types and vintages in dozens of locations. Project description: Resources are needed for running building energy models for estimating the energy impacts of new technologies. Hundreds of prototype buildings covering multiple building types and vintages and in many locations of the country are required. This will utilize NREL's OpenStudio, an Open Source building energy modeling package. Computed data will be post processed using Python. Industry partnership: Project URL: Requested allocation: 3000 Q1: 0 Q2: 1000 Q3: 1000 Q4: 1000 Justification: N/A Storage requirements: under 1 TB 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 allocations-admins(a)lcrc.anl.gov<mailto:[email protected]> https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins _______________________________________________ allocations-admins mailing list allocations-admins(a)lcrc.anl.gov<mailto:[email protected]> https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

[LCRC Accounts] Project Request: KTP
by accounts＠lcrc.anl.gov 30 Jan '17

30 Jan '17

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: Ahren Jasper Applicant's institution: ANL Applicant's division: CSE Project Name: KTP Project title: Temperature- and Pressure-Dependent Kinetics Associated funding: DOE BES Other Systems: Chemistry's Blues cores b431-b460 Chemistry's linux 1300-core cluster "Linus" Science: In recent work, we have shown that quantitative predictions of gas phase chemistry can be made using a combination of classical trajectories (molecular dynamics), transition state theory, and master equation calculations. The present project focuses on using classical trajectories to predict energy transfer parameters for use in master equation calculations for real fuels of relevance to combustion. Whereas in past work, we have considered small test case fuels, such as methane, real fuels often contain much larger species such as dodecane and the large esters that appear in biodiesels. The main goal of this project is to use high performance computing to adapt our methods to the treatment of larger (real) fuels. Project description: We will (1) fit full-dimensional potential surfaces, and (2) run trajectories to calculate energy transfer. Step (1) involves parallel sampling of the intermolecular potential (which has a dimensionality of 3, 5, or 6 for atomic, diatomic, and polyatomic baths, respectively) using commercial quantum chemistry (QC) programs (principally Molpro). "Good" QC methods, such as counterpoise corrected CCSD(T)/CBS, can be prohibitively expensive, even in large parallel computing environments. We will therefore continue our work optimizing (a) the choice of QC method (considering MP2, dispersion corrected DFT, etc) and (b) sampling point placement (considering biased distributions and quasirandom (Sobol) sampling). In preliminary work, we have found (a) that MP2 is often accurate for Ar as a bath gas, and (b) a biased quasirandom sampling can increase efficiency by a factor of 15. Still, for large molecules, sampling will be expensive, requiring ~10000 1-hour QC calculations. Fitting the resulting data will be done using a genetic algorithm or linear least squares fitting. The cost of the fi tting is negligible relative to the cost of sampling. Step (2) involves using our parallel trajectory code DiNT to run energy transfer trajectories. 10,000 independent trajectories are required to characterize energy transfer for each system of interest, with each trajectory taking 1-60 minutes, depending on the size of the system. In this first step, we will consider medium sized alcohols, peroxides, and esters. For each of these three class, we will consider 9 cases (e.g., the three alcohols ethanol, butanol, and hexanol for each of the three baths He, Ar, and N2) for a total of 27 cases. The cost of both steps 1 and 2 will scale dramatically (in different ways) based on system size, with the largest systems considered requiring most of the computer time. For example, we estimate 10000 samples of the hexanol+Ar potential to require 5000 hrs, though we likely have to "waste" processors due to I/O limitations, thus increasing the core-hour cost by up to a factor of 3. Altogether, the quantum chemistry costs should require: 3 classes * 9 cases/class * 10,000 core-hours/case = 270,000 core-hours. In the second step, for each of the 27 cases, we will run 10,000 trajectories. These take 30 minutes, on average, for a total of 135,000 core-hours. The trajectory calculations run at near 100% parallel efficiency and scaling. Steps 1 and 2 thus require a total of 405,000 core-hours, to which we add 20% for testing and development, leading to a total request of 486,000 core-hours. Industry partnership: None Project URL: Requested allocation: 486000 Q1: 0 Q2: 86000 Q3: 200000 Q4: 200000 Justification: Storage requirements: Less than 1 TB 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

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 30 Jan '17

30 Jan '17

Hello, A change in allocation has been requested: Requester: neveu (Nicole Neveu) Project: AWA-beam-dynamics Title: High Accuracy Beam Dynamic Simulations Description: The primary goal of this project is to simulate the two linear electron accelerators at the Argonne Wakefield Accelerator facility. To date, a full 3D models of the beam lines has not been simulated. Using OPAL-T and the parallel capabilities of Blues, we seek better agreement between simulations and experiment by accounting for 3D effects, space charge, coherent synchrotron radiation, and wakefields. Current: undetermined amount Justification: Requested: 15000 A specific reason has been given: The parameter space for one of our current projects is bigger than expected. A large random sample was done to characterize the space. This took the remainder of our hours. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0