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

[LCRC Accounts] Project Request: PEMFC-Microstructure
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '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: Rajesh Ahluwalia Applicant's institution: ANL Applicant's division: NE Project Name: PEMFC-Microstructure Project title: Nano-scale Characterization and Electrochemical Transport in Electrodes of Polymer Electrolyte Fuel Cells Associated funding: DOE-EERE-FCTO Other Systems: Limited availability on NE cluster Eddy: 48 nodes, 8 cores/node, 768 GB memory Limited availability on NE cluster Eddy2: 8 nodes, 32 cores/node, 1028 GB memory Limited availability on TRACC clusters Zephyr and Phoenix Science: Resolve detailed electrode microstructure at nanometer (nm) length scale in terms of spatial and size distributions of catalyst, carbon, ionomer, primary pores and secondary pores Conduct multiphase simulations of electrochemical kinetic and transport processes on the nanoscale microstructure Develop foundational understanding of the relationship between electrode performance and its microstructure Project description: We are developing a new approach to quantitatively characterize the detailed cathode electrode microstructure from nm to μm length scales by combining data from different experimental techniques: nano-scale X-ray computed tomography (X-CT) in phase contrast mode at APS 26-ID-C and 32-ID-C beam lines for secondary pores and solids (~20 nm resolution); X-CT in absorption contrast mode for carbon and ionomer distribution within the solid phase (~20 nm resolution); transmission electron microscopy (TEM) and X-ray scattering in ultra-small angle (USAXS) to small-angle (SAXS) regions at APS 9-ID-C for primary carbon and catalyst particle size distributions (~1 nm resolution); and BET and mercury intrusion porosimetry (MIP) for primary pore size distribution. We have developed a stochastic percolation algorithm in MATLAB to incorporate the experimental data from the above characterization techniques with different length scales in one geometric representation (hybrid microstructure) with the required resolution determined by the technique with the highest resolution. We have run COMSOL and STAR-CCM+ simulations on the hybrid microstructure to determine anisotropic oxygen transport and proton conductivity on extracted sample volumes. Because of limited computer resources, the simulations have been limited to 1 um x 1 um x 1 um domain, discretized in 64 million (64M) nodes for 2.5 nm length scale resolution. We have started to develop pore scale microscopic models in MATLAB and C-language for liquid water transport in the hybrid microstructure. We have been successful in developing algorithms for the dynamic invasion model (DIM) and obtaining initial illustrative results on coarse grids. Meaningful results will require access to full parallel computing on a supercomputer. Proposed Work on LCRC 1. Expand hybrid microstructure model to combine information from phase contrast and absorption contrast • Refine ionomer structure, ionomer film thickness, and ionomer segregation • Validate results with TEM data 2. Conduct pore-scale DIM simulations on full thickness of electrodes • 1 um x 1 um x 10 um domain, discretized using 640M nodes for 2.5 nm resolution and 10B nodes for 1 nm resolution • Extract transport properties for use in macroscopic model 3. Extend DIM to water transport in diffusion media (DM) • Conduct liquid water transport in cathodes attached to macro porous layer (MPL) with gas diffusion layer (GDL) • Extract transport properties for use in macroscopic model • Validate results with nano and micro X-CT data 4. Combine DIM with pore-scale oxygen/water vapor transport in electrode pores and ionomer 5. Create foundational understanding of electrochemistry and transport processes by conducting simulations of oxygen reduction reaction on catalyst particles combined with gas and liquid transport in hybrid microstructure Project Team Members: Rajesh K Ahluwalia (NE), Cankur F Cetinbas (NE), J-K Peng (NE), Xiaohua Wang (NE), Dennis. D. Papadias (NE) On-going Experiments: Nano-XCT studies of electrode microstructure at APS; USAXS and SAXS studies of inks and electrodes at APS; TEM studies of cathodes at ORNL Industry partnership: Project URL: http://www.fcpad.org Requested allocation: 400000 Q1: 100000 Q2: 100000 Q3: 100000 Q4: 100000 Justification: Storage requirements: 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] Yearly Allocation Request for Atmos_Chem
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Rao Kotamarthi Project Name: Atmos_Chem Division: EVS Project title: Process Scale Modeling of Atmospheric Chemistry and Transport Associated funding: DOE,DoD Other Systems: NERSC Science: A primary objective is to perform dynamic downscaling of climate model simulation results for regional scale applications. In particular, we will focus on the uncertainties in downscaled projections resulting from boundary conditions, the utility of spectral nudging and grid scale nudging for improving the energy balance of regional scale climate models over long integration times. In particular, we are exploring the development of new regional scale climate models as there is a need to develop a code base that bridges the scales from mesoscale to microscale using continuous mesh refinement that can allow for local refinement of grids from kilometers and scales with well established physics parameterization to meters, where the flow is completely resolved without the need of nesting. Project description: There are several existing mesh geometries and dynamical cores that allow continuously refined meshes, for example, the Model for Prediction Across Scales (MPAS) develop by Los Alamos National Laboratory and the National Center for Atmospheric Research (NCAR); and the High-Order Method Modeling Environment (HOMME) contributed by NCAR, DOE laboratories and universities. These models are built using various resolution meshes which have smoothly-varying mesh transitions. This capability ameliorates many issues associated with the traditional mesh refinement strategy of one-way and two-way grid nesting where the transitions are abrupt. However, currently, both of these dynamic cores are only used in global scale models. It is well known that running global scale models at very high resolutions is computationally expensive. For example, running a uniform 3-km mesh over globe contains 65.5 millions of horizontal grid cells. Tests on NCAR’s Yellowstone found that the uniform 3-km mesh with 41 layers requires around 190000 core-hours per simulated day; while BG/Q hardware at Argonne tends to require around five times the number of core-hours as Yellowstone (personal conversation with Michael Duda, NCAR). However, if one is only interested in resolving flow and atmospheric dynamics over a limited area of the earth system, then resolving flow in remote regions from the boundary of the limited area of interest is not necessary or computationally relevant. Thus, there is an urgent need to develop an exascale capable code for atmospheric simulations over limited regions based on the same or similar dynamical core as MPAS and/or HOMME. Doing so will only focus on the area of interest and allow continuous refinement from kilometers to meters with much cheaper computational resources. For this proposal the primary activity is developing a process for prescribing the boundary conditions to the limited area model (LAM) with unstructured grids. Three types of lateral boundary conditions will be tested: (a) Smooth flow boundary conditions, which fixes the velocity (U,V) at the boundary for the model simulation period; (b) Cyclic boundary conditions, which cycles a time varying boundary condition over a certain period of the model simulation (for example a day) over the entire simulation period (for example a month); and (c) variable (observational) boundary conditions, which provide temporal varying boundary conditions for the model. The terrain following coordinate or the hybrid coordinate used in the MPAS model will be used as the surface boundary in the LAM. The lateral boundaries will be modeled using a similar approach used in the fixed mesh mesoscale meteorological models, such as the WRF model, by defining a buffer region extending over several grid po ints and using interpolation of the flow fields to the buffer zone. Industry partnership: None Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 200000 Q1: 50000 Q2: 50000 Q3: 50000 Q4: 50000 Justification: Storage requirements: 1TB Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request for Aerosols-Climate
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Rao Kotamarthi Project Name: Aerosols-Climate Division: ER Project title: Modeling the Impact of Greenhouse Gases on Radiative Forcing in Regional Atmospheric Simulation Associated funding: DOE Other Systems: NERSC, NCSC Science: The impact of non-CO2 greenhouse gases such as methane, nitrous oxide and aerosols on radiative forcing is investigated using numerical models. Project description: This DOE Biological and Environmental Research (BER) division Atmospheric System Research funded, Science Focus Area (SFA) project seeks to improve fundamental understanding of small-scale processes and to reduce key uncertainties in aerosols, clouds, and aerosol-cloud interactions. The SFA will have aerosol and cloud themes. The aerosol theme is organized around the chemical, physical, and optical properties of atmospheric aerosols, as well around the formation, growth, and removal phase of the aerosol life cycle. The cloud theme focuses on improving the representation of the interactions between cloud-scale physics and dynamics with the environment for marine boundary layer clouds, mixed-phase clouds, and deep convective clouds. Aerosol theme knowledge gaps include reducing uncertainties in global predictions of aerosol radiative forcing; identifying key nucleation and growth mechanisms responsible for new particle formation and improving their rep resentation in models; and achieving a quantitative understanding of the processes controlling aerosol and cloud condensation nuclei populations in the marine boundary layer. Cloud theme knowledge gaps include determining the net effect of coupled processes including turbulence, drizzle, and aerosol-cloud interactions on marine boundary layer cloud properties; improving model representation of the partitioning of super-cooled liquid water and ice in arctic mixed-phase clouds; and determining how and which processes determine the properties of deep convective cores so that they can be better represented in models. We will be using global, regional and process scale models to address these science gaps during the course of the project. Industry partnership: None Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 200000 Q1: 50000 Q2: 50000 Q3: 50000 Q4: 50000 Justification: Storage requirements: 1TB Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request for chargetransport
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Marton Voros Project Name: chargetransport Division: MSD Project title: Charge transport in energy materials from first principles Associated funding: LDRD -- Charge Transport in Nanostructured Materials from ab initio Simulations Other Systems: NERSC, 1.000.000 hours (project on nanoparticle solar cells) Science: The ultimate goal of this project is to understand and engineer charge transfer processes in materials relevant for energy storage and conversion. Following the success of this project in the past years on understanding transport in lithium manganese oxide (LMO) based batteries (one paper in revision in Nature and another being close to submission), this year we will extend our investigation to other cathode materials relevant for lithium-ion batteries. In particular, using our recently developed constrained density functional theory code (J. Chem. Theory Comput. 13, 2581 (2017)), we will model charge transport in layered lithium cobalt oxide (LCO), a paradigmatic material for Li-ion batteries. In the case of favorable results, and if time permits, we will also investigate alloyed lithium transition metal oxides. Since this effect has been ignored in all computational studies to date, we believe our investigation could open new doors in the design of better cathodes . In particular, we will be able to investigate how lithium and electron transfer are coupled and how the movement of the electronic charge affects ion transport. Such electron-ion coupled processes (e.g. proton-coupled electron transfer) are known to be imperative to understand the catalytic behavior of organic or organometallic materials but this effect has been largely ignored in the lithium-ion battery community. In parallel, we will start modeling photo-induced processes using time-dependent density functional theory. In our paper under revision, we have experimentally show that LMO is able to absorb visible light, which in turn gave rise to enhanced charge and ion transport. However, it is not clear what the electronic structure origin of the absorption peaks is so the mechanistic reasons for the enhanced charge and ion transport was only speculated. At the end of this allocation year, we will do the initial steps towards understanding this process and we will start modeling the photo-absorption behavior of LMO and LCO. Project description: We will be using first principles methods based on density functional theory. In particular, for simulating oxide cathode materials we will be using the plane wave code Quantum Espresso (QE). QE can make use of efficient linear algebra (Lapack, BLAS, Scalapack, BLACS) and FFT libraries, if they are available. QE was shown to efficiently scale to a few hundreds of processors in the case of non-hybrid functional calculations. Hybrid functional calculations can scale up to a few thousands of cores, especially because of recent methodological and algorithmic development (J. Chem. Theory Comput., 2012, 2242 (2016)). NERSC staff has restructured the algorithm behind hybrid functionals in Quantum-Espresso and has shown that this method is capable of making use of the new Cori KNL nodes due to its hierarchical, multi-level parallelization strategy that involves using several layers of MPI and OpenMP parallelization (e.g. http://www.nersc.gov/users/computational- systems/cori/application-porting-and-performance/application-case-studies/quantum-espresso-exact-exchange-case-study/). We expect the same or similar benefits when using the KNL nodes of the new Bebop cluster for the hybrid functional calculations, however, we know from our experience that Quantum-Espresso also runs fine the Blues cluster and we expect it can also nicely run on the broadwell nodes. Indeed, our initial tests on small systems during the early user access program showed good performance. We will plan to use 60% of our allocation on non-hybrid functional calculations and we will use the remaining 40% of the time to do more expensive (but also more accurate) hybrid functional calculations on selected systems. A typical non-hybrid functional calculation will take several hours of wall clock time (up to tens of hours) on a few hundred processors. There is an efficient restart mechanism in QE allowing us to run smaller jobs. This is useful if the wall-clock time limit does not allow to finish big calculations in one step. To simulate charge transfer processes, we will first identify sites that localize charges in the form of polaron-like carriers. Then, we will use nudged-elastic band and our own implementation of constrained DFT (CDFT) to simulate charge transfer between the sites. This will allow us to compute the parameters entering semi-empirical charge transport models, such as Marcus-theory. Our own implementation of CDFT is integrated into a private version of the most recent release of Quantum-Espresso and it inherits its performance, although typical CDFT calculations are about one order of magnitude more expensive since they require a double self-consistent cycle instead of the regular single self-consistent cycles. For the optical calculations, we will use time-dependent density functional theory as implemented in the Quantum-Espresso code and first we will be looking lithium cobalt oxide which does not require hybrid functionals of Hubbard corrections to get the correct ground state. Thus, we will expect we will be able to use a low level of theory, which is necessary to make the calculations doable. For the charge transport subproject, we expect to run about 100 regular non-hybrid DFT calculations on LCO, each on ~256 cores (assuming KNL nodes) that would last several hours, possibly in the order of 4-5 hours. This would allow us to have a quick estimate for most relevant charge transfer pathways. Then, after picking the most important configurations (<20), we will run CDFT calculations that would last 10-24 hours on 256 cores. We will then repeat a few calculations using hybrid functionals on ~1024 cores. We estimate we will run about 10 calculations that would last 24 hours. Optical calculations using time-dependent density functional will be run on the perfect bulk material and we estimate its cost about 100 times of a regular non-hybrid ground state. All in all, we request 700.000 core hours for the entire year. Since more expensive calculations will require performing less expensive calculations first, we request proportionally more time for the third and fourth qua rter. Finally, for the reasons discussed above, we believe that it is only our hybrid functional calculations that can really benefit from using the KNL nodes, however, we can definitely use the broadwell nodes efficiently and we had a successful experience in using Quantum-Espresso of Blues. Industry partnership: Project URL: Current FY Hours Used: undetermined amount New FY Requested allocation: 700000 Q1: 125000 Q2: 125000 Q3: 225000 Q4: 225000 Justification: This is detailed in the Project Description: hybrid functional calculations have been shown by NERSC staff to scale up to thousands of cores due to recent numerical and algorithmic advances (J. Chem. Theory Comput., 2012, 2242 (2016), http://www.nersc.gov/users/computational-systems/cori/application-porting-a…) we thus expect that calculations using 40% of our requested allocation can scale up to thousands of cores. Since this part of Quantum-Espresso makes use of multi-level, hierarchical parallelization, if necessary, we will perform benchmarks to find the ideal parallelization layout. Storage requirements: 2 TB. We will need to, at least temporarily, save wave functions and charge densities to enable efficient restarts of QE calculations. Since these quantities are saved on grids and can be several tens of GBs in size, we estimate we will need about 2 TB of storage. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request for QM_Storage
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Rajeev Surendran Assary Project Name: QM_Storage Division: MSD Project title: Organic Materials for Energy Storage Associated funding: BES: JCESR Other Systems: CNM Science: Understanding, controlling, and eventually manipulating functional groups on promising organic molecules can be used to obtain the wide electrochemical window needed for the next generation of non-aqueous redox flow battery electrolytes. Combined electrochemical measurements in combination with accurate quantum calculations to probe structure-activity relationships of redox active molecular (RAM) species and to study their interactions with the surrounding electrolyte with an overall goal of developing key descriptors for predicting performance and stability of various organic compounds is central to this project. Predictive quantum chemical simulations based on density functional theory is utilized to compliment electrochemical experiments to investigate the redox properties of redox active organic molecules. The project is an integral part of the Materials Discovery associated with the Non-aqueous redox flow thrust of Joint Center of Electrical Energy Storage (JCE SR) at the Argonne National Laboratory. Project description: Computations based on first principles (Gaussian 09, quantum espresso, and CPMD) will be performed. Utilizing the power of accurate first principle simulations is essential to gain deeper electrochemical-mechanistic understanding and accelerate discovery. Simulations and models that provide effects of the dynamics of solvation, external potentials, impurities, reaction intermediates is essential to capture the critical parameters that dictate the reactivity and longer term stability. The quantitative understanding of the critical parameters would enable the development of electrochemical structure activity relationships, optimize electrochemical conditions and to provide insights into the reactivity and stability of a diverse material dataset. First, we identify fiducial systems including the most promising catholyte and anolyte molecular system with experimental inputs and generate a snapshot of the nano-dimensional description via ab initio dynamic simulations (AIMD). Multiple snapshots or cases representing distinct scenario will be utilized to further to obtain better s ampling of the conditions and compositional variance. Second, for each case the impact and the roles of critical parameters (dielectric medium, concentration of neutral/charged species, external potential, nature of the electrode, and intrinsic reactivity) will be assessed and quantified.. Dependency between selected critical features will be investigated in detail (e.g.: the effect of concentration/electrode/external potential for a specific chemical reaction). Simulations will be performed to estimate the rate of polymerization, electrode reaction, and likely reaction with intermediates/impurities, self-exchange of the redox-mers. Finally, these computations enables a library of fundamental knowledge (via publications) connecting the electronic and environmental structure in the liquid phases and in the surfaces with quantitative understanding of the critical parameters. This can be utilized to guide experiments and develop efficient screening to accelerate discovery. Industry partnership: None: Project URL: http://www.jcesr.org/research/redox-flow/ Current FY Hours Used: undetermined amount New FY Requested allocation: 500000 Q1: 125000 Q2: 125000 Q3: 125000 Q4: 125000 Justification: Atomistic calculations using G09 is computationally less demanding and performed on 1 node in BEBOP. Quantum espresso, latest version, is shows good parallelization up to 8 nodes (288 processors). The request of 500K core hours is based on our previous experience in cluster, surface and dynamics calculations. Storage requirements: 1TB Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Request: DeepMic
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '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: Eric Schwenker Applicant's institution: ANL Applicant's division: NST Project Name: DeepMic Project title: Deep Learning Atomistic Structure from Microscopy Images Associated funding: LDRD Other Systems: JLSE, Carbon (100,000 core-hours/year), NERSC (250,000 core-hours/year) Science: In the present work, we are developing a framework for experimentally-driven atomic structure optimization by combining atomistic modeling with electron microscopy image simulation and deep learning. Existing studies that integrate deep learning techniques into microscopy treat content analysis as a general classification problem, attempting to assign classification labels such as austenitic, ferritic, or martensitic, for example, to a micrograph of steel. While this provides a useful framework for organizing large datasets of specific microscopy images, it fails to use the content in a generative way (i.e. to inform model creation or update an iterative optimization scheme). In order to use atomic resolution images for such tasks, it is critical that image feature detection be carried out in a robust manner. Currently the image features are extracted using a hand-crafted peak detection methods, which limits the types of microscopy images that thi s representation can be applied to. A wealth of open source deep learning approaches to automatic feature extraction exist, and we believe they could be leveraged to automate microscopy interpretation in the design of physical models and microscopy image analysis and retrieval across large datasets. Project description: Deep neural networks (DNN’s) have been shown to be able to capture high-level features describing image data, even when trained on datasets of unrelated objects (i.e. faces or natural scenes). This is attributed to the notion that the features learned are fundamental to visual recognition, and are thus transferable across all images in the general sense. In this project we propose to leverage pre-trained DNN architectures such as YOLO [J. Redmon, CVPR (2016)], DeepMask and SharpMask [P. Pinheiro et al. NIPS (2015), ECCV (2016)], and inception-v3 [C. Szegedy, arxiv (2015)] (pre-trained for other recognition tasks) to test their accuracy in identifying the atomic columns present over a dataset of several thousand simulated atomic resolution microscopy images that are subject to varying levels of synthetic imaging noise. Additionally, we will use these pre-trained DNN architectures, as the basis for “fine-tuning” new feature detection models on si mulated microscopy images. Our (Eric Schwenker, Maria Chan) previous efforts in this area have led to the development of a framework for matching atomic resolution microscopy images. There is a noticeable decline in matching performance as the images become increasingly corrupted with noise and blur. The belief is that the matching performance is not significantly hindered by the metric used for matching, but rather the inability to consistently identify the atomic columns across all the degraded images. Our goal is to use the automatic feature extraction tools from deep learning to improve the image matching framework, and then show that this can be applied to both experimentally-driven atomic structure optimization, and across different imaging modalities to quantify similarities between images. The breakdown of the required calculations is as follows: (1) Image simulation and labelling for a complete set of ICSD prototypes: Currently, the Inorganic Crystal Structure Database (ICSD) contains ~9,000 unique crystal structure prototypes. We will construct an initial set size of ~36,000 pristine simulated microscopy images (~ 4 images per structure prototype) where the image peaks (atomic columns) are labeled automatically based on the projection of the structure on the imaging plane. In addition, we will create 2 complete datasets of corresponding image matches that are corrupted with varying levels of simulated imaging noise. With this, the entire dataset will total (9000 x 4) x 3 = 108,000 images. To simulate these images, a new electron scattering simulation algorithm (PRISM) [C. Ophus, Adv. Struct. Chem. Imaging (2017)], will be used. With the settings we have determined for the algorithm, a single image requires 2 hours of simulation time on a single core. With ~108,000 images, these calculations estimated to take 216,000 core hours. An additional 28,000 core hours will be used to evaluate the accuracy of 3 separate pre-trained DNN detection models. TOTAL CALCULATION TIME: 244,000 core hours (2) Fine-tuning pre-trained DNN’s on ICSD dataset: Training DNN’s from scratch is a laborious process that can take several weeks on modern computing platforms. To avoid this, we will “fine-tune” the existing pre-trained DNN architectures using their respective open-source computing frameworks: darknet (http://pjreddie.com/darknet) for YOLO, Torch (http://torch.ch/) for Deep/SharpMask, and TesorFlow (https://www.tensorflow.org/) for inception-v3 . This involves adjusting the weights of the later layers that are more specific to the details of the microscopy images themselves. For this, we will allot ~ 60,000 core hours for each of the three existing DNN architectures explored. TOTAL CALCULATION TIME: 180,000 core hours (3) Image matching for atomic resolution datasets: The current image matching framework is setup to test matching performance on a dataset of ~25,000 simulated microscopy images A full performance run of a single DNN detector over the entire set will require ~20,000 core hours. The 3 pre-trained DNN detectors, along with the 3 fine-tuned DNN’s will require 6 x 12,000 = 72,000 core hours to evaluate. TOTAL CALCULATION TIME: 120,000 core hours Industry partnership: Project URL: Requested allocation: 496000 Q1: 124000 Q2: 124000 Q3: 124000 Q4: 124000 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

1 0

[LCRC Accounts] Yearly Allocation Request for UNEDF_Optimization
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

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

1 0

[LCRC Accounts] Yearly Allocation Request for ACPC
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Scott Collis Project Name: ACPC Division: EVS Project title: Understanding storm tracks near Houston in support of the Aerosols, Clouds, Precipitation and Climate initiative Associated funding: DoE CESD: ARM CMDV. Other Systems: None Science: The Aerosols, Clouds, Precipitation and Climate aims to look at the evolution of storms as the move from pristine to polluted environments. The team is proposing a field campaign in Houston, TX. This very short project will process and map 15 years of NEXRAD data from the area and perform cell tracking to derive a climatology of storm tracks. This will be used in site-ing decisions and will also form a neat publishable project in as of itself. Project description: This renewal of our previous project will build on work performed in the exploratory 2017 project. We will extend our use of the Dask module to use more cores and process the rest of the five year dual polarimetric radar record. Radar volumes will be processed using the Python ARM Radar Toolkit. This includes quality control, calibration and the retrieval of specific differential phase. Following this processing data will be mapped to a model like grid in preparation for cell tracking. Once the CF complaint radar grids are compiled our tracking code, TINT (TINT Is Not TITAN) will be used to extract the cell locations, tracks and new, enhanced, cell statistics. We have shown these processes on Beebop can scale to 1,200 cores. Approximately half the allocation will be used for processing and half for analysis. Industry partnership: None Project URL: http://asr.science.energy.gov/meetings/stm/2017/presentations/380.pdf Current FY Hours Used: undetermined amount New FY Requested allocation: 100000 Q1: 25000 Q2: 25000 Q3: 25000 Q4: 25000 Justification: Storage requirements: We have our own Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Yearly Allocation Request for Cat_Biomass
by accounts＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Rajeev Surendran Assary Project Name: Cat_Biomass Division: MSD Project title: Computational Studies of Biomass Catalysis Associated funding: EERE Other Systems: CNM Argonne, Blue Gene/P Argonne, NERSC Science: This computational project will be performed to gain fundamental understanding of the chemistry of biomass reactions and help to design catalysts that improve the efficiency of biomass conversion to transportation fuels or industrial chemicals. The project is an integral part of the computational modeling as part of the Computational Biomass Pyrolysis consortium funded by Energy Efficiency and Renewable Energy (EERE). Efficient chemical transformation of biomass is essential to produce sustainable energy and industrial chemicals. Conversion of biomass to useful chemicals include sequence of chemical transformation including, C-O bond cleavage and C-C bond formation. Project description: We employ first principles-based calculations (density functional theory and other accurate methods) to understand the kinetics and thermodynamics of catalytic reactions. The computational modeling include: Cluster calculations, surfaces and large scale dynamic simulations. This project involves mainly application of existing computational modeling techniques to explore the chemistry and discovery of catalysts. The methods employed are state of the art and accurate to make in situ prediction possible. The computational chemistry approach is based on the accurate principles such as density functional theory and ab initio methods. Software packages such as Gaussian 09, Quantum Espresso, and CPMD were the main tools employed for this information. Following specific problems will be addressed using first principle-based simulations. Recently our group has shown that Mo2C catalysts are active catalysts for the deoxygenation of model biooil compounds. However, regeneration of the Mo2C catalysts by oxygen desorption was shown to be slow so catalysts with faster oxygen desorption properties are sought. Mo2C is a good deoxygenation catalyst due to the oxophilic nature of molybdenum making strong Mo-O bonds upon deoxygenation. Therefore, other oxophilic early metal carbides could be potential catalysts. As MoN has also shown deoxygenation properties, early metal nitrides could also be potential catalysts. Specifically four metal carbides and five metal nitrides were chosen as catalysts: WC, Nb2C, V2C, TiC, MoN, VN, WN2, NbN, and TiN. These polymorphs were chosen as they are the most stable polymorphs of each set of elements with the exception of the Group V metal carbides (V, Nb). For these systems the most stable polymorph (M6C5) is too large and would be too computationally expensive; so, the second most stable polymorph was selected. As nickel dopants have shown to increase catalytic activity, ternary Mo-Ni-X systems were also considered in the study (Ni2Mo4C and Ni2Mo3N). The selected catalysts will be screened computationally for potential as deoxygenation catalysts and compared with known Mo2C catalysts. The screening study will focus on 3 properties to determine the best deoxygenation catalysts: O* and C* binding energies at various O* and C* coverages, and the kinetics for O* desorption as water. The O* and C* binding energies will provide crucial information about the nature of the surface under reaction conditions. Also the binding energies would identify the most thermodynamically favorable reaction sites for deoxygenation and serve as a descriptor to compare the thermodynamics of deoxygenation. The thermodynamics of deoxygenation must be balanced against regeneration of the catalyst as large O* binding energies may suggest that regeneration of the catalyst could be difficult. Therefore the kinetics of desorption of bound oxygen as water will be studied to ensure no kinetic barrier to regeneration. The ideal catalyst would bind oxygen strongly, but have a low barrier to water formation. Using the properties determined in the initial screening, catalysts will be chosen for a full mechanistic probe of the reaction of ethanol over the catalyst surface to test our catalyst selection criteria. As such catalysts predicted to be both good and bad by the initial criteria will be considered. From the mechanistic study important information on selectivity (C-O vs C-C bond cleavage) and kinetics for biooil upgrading will be determined. The barriers and selectivities determined for the catalysts tested will be compared with properties of the intermediates and from the initial screening to determine potential descriptors for the reactions. With these descriptors our initial catalyst screening criteria can be improved for testing on more carbides and nitrides. Additionally, we plan to Investigate competitive adsorption of water and ketones (4-heptanone, 2-pentanone) on selected metal oxide catalysts (TiO2, MgO, ZrO2, and MgZrO) and develop reactivity descriptors of ketones in the presence of varying concentration of water. NREL will investigate the relative adsorption of ketones and water and develop co-adsorption models and phase diagrams. ANL will develop reactivity descriptors and mechanisms of the aldol reaction at the surface using atomistic simulations. The results will inform experimental efforts of CUBI (WBS#) and CFP (WBS#) to optimize the influence of water in the metal oxide catalysis. Industry partnership: Industry Partnership: Project URL: http://cpcbiomass.org/ Project URL: http://cpcbiomass.org/ Current FY Hours Used: undetermined amount New FY Requested allocation: 1000000 Q1: 250000 Q2: 250000 Q3: 250000 Q4: 250000 Justification: Large allocation requests are due to the following reasons: 1. We plan to perform Ab initio screening of nitride catalyst using VASP. This is followed by transition state calculations using the NEB that demands significant computational resources. 2. In the Nitride catalyst design and competitive water adsorption project, we are planning to perform Ab Initio Dynamics simulations using CPMD or Quantum Espresso. This is also computational demanding. Storage requirements: 1TB Thank You, The LCRC Accounts System

1 0

New LCRC report submitted for UNEDF_Optimization
by support＠lcrc.anl.gov 07 Sep '17

07 Sep '17

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2017/?project=UNEDF_Optimiz…

1 0