January 2016 - Allocations-admins - mailman3.cels.anl.gov

Re: [allocations-admins] Project Allocation Request - NEXTGENOPT
by Roberts, John E. 22 Jan '16

22 Jan '16

The 160K hours have been granted for this project. -- John Roberts Argonne National Laboratory CELS Systems jeroberts(a)anl.gov On 1/22/16, 9:41 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: >Lets approve the request for 160K. > >Ray > > > >On 1/21/16, 3:45 PM, "Kim, Kibaek" <kimk(a)anl.gov> wrote: > >>Dear Ray: >> >>Here is my estimate: >> >>1. A research project that is currently using resources will further >>require 160000 core-hours in a month. >>2. Another research project that will start running computations in >>February is expected to require 50000 core-hours. >> >>So, I would need nearly 160000 core-hours more by the end of March. >> >>Best, >>Kibaek >> >>> On Jan 21, 2016, at 9:34 AM, Bair, Raymond A. <rbair(a)anl.gov> wrote: >>> >>> Dear Kibaek Kim, >>> >>> We need your feedback to proceed. >>> >>> Regards, >>> >>> Ray >>> >>> >>> >>> On 1/14/16, 4:21 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: >>> >>>> Dear Kibaek, >>>> >>>> Could you estimate how much more time you now and the end of March? In >>>> March the LCRC Allocations Committee meets to make decisions about the >>>> time allocations for Q3 and Q4 (April-Sept.). So the immediate >>>>decision >>>> is between now and March 31st, and the rest of your request can be >>>>added >>>> to your original request for Q3 and Q4. >>>> >>>> Regards, >>>> >>>> Ray >>>> >>>> >>>> ----------------------------------------- >>>> Ray Bair >>>> Argonne National Laboratory >>>> and the University of Chicago >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> On 1/14/16, 10:00 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: kibaekkim (Kibaek Kim) >>>>> Project: NEXTGENOPT >>>>> Title: Next generation optimization >>>>> Description: Sponsored by a Department of Energy (DOE) Early Career >>>>> Award, we are currently developing new algorithms for UQ/SO that have >>>>>a >>>>> holistic view of the entire data-modeling-optimization process. A >>>>> particular example of such algorithms is a new clustering-based >>>>> interior-point strategy for stochastic programs in which scenarios are >>>>> compressed based on their influence on the first-stage decision, not >>>>>on >>>>> data. This has shown to lead to drastic compression rates (and >>>>>solution >>>>> times) of above 80%, which cannot be achieved with data compression >>>>> alone (see Figure 1). The compression is done adaptively, at the >>>>>linear >>>>> algebra level, by looking at the different contributions of the >>>>> scenarios on the Schur complement. >>>>> >>>>> We are also currently developing new interior-point algorithms for >>>>> nonconvex structured optimization capable of dealing with complex >>>>> multi-scale physics (typically involving large networks of partial >>>>> differential equations) and stochastic components. These problems >>>>>arise, >>>>> for instance, from natural gas inventory (line-pack) management (see >>>>> Figure 2). This requires of a redesign of existing algorithms and >>>>> modeling environments capable of conveying problem structure down to >>>>>the >>>>> linear algebra kernels. In addition, hybrid linear algebra kernels >>>>>need >>>>> to be designed to exploit different structural facets. Problems with >>>>> billions of variables are being targeted. >>>>> Current: undetermined amount >>>>> Justification: >>>>> >>>>> Requested: 400000 >>>>> >>>>> A specific reason has been given: >>>>> We are running a large number of optimization instances for a power >>>>>grid >>>>> application. The application analyzes the cost efficiency of the large >>>>> power grid system in California with uncertain wind power production. >>>>>As >>>>> a result, a large scale stochastic optimization problem needs to be >>>>> solved with different seasons and different parameter settings, which >>>>> requires more core-hours to run such optimization problems. >>>>> >>>>> 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 >>> >> > >_______________________________________________ >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 - dmft_for_oxides
by Bair, Raymond A. 22 Jan '16

22 Jan '16

Dear Hyowon Park, Could you estimate how much more time you need between now and the end of March? In March the LCRC Allocations Committee meets to make decisions about the time allocations for Q3 and Q4 (April-Sept.). So the immediate decision is between now and March 31st, and the rest of your request can be added to your original request for Q3 and Q4. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/22/16, 11:06 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: hyowon (Hyowon Park) > Project: dmft_for_oxides > Title: DFT+DMFT Calculations of Transition Metal Oxides > Description: A DFT+DMFT code consists of a DFT part and a DMFT part. >Most of modern DFT codes are efficiently parallelized such that >simulations of large supercells can be performed. A DMFT calculation >requires an accurate convergence of the quantum Monte Carlo method. The >quantum Monte Carlo simulation is scalable almost linearly as the number >of cpus. These codes can be compiled using the Intel compilers and MPI >which are already installed in LCRC machines. >Therefore, the parallel calculation using the clusters in LCRC will be >inevitable for performing precise and efficient DFT+DMFT simulations. > >During FY 2015-16, I will perform DFT+DMFT calculations to study an >accurate ground-state description of strongly correlated materials >including structure, total energy, and magnetism as well as excited-state >properties such as the one-particle spectral function and two-particle >magnetic/charge response functions. Specifically, I will study the strong >spin-lattice-charge coupling in La1/3Sr2/3FeO3 oxides, the magnetic >susceptibility in CePd3 heavy fermions, the spin-state transition in >LaCoO3, the metal-insulator transition in rare-earth nickelate thin >films/superlattices, and the energetics of UO2 heavy fermions. These >DFT+DMFT outputs can be directly compared to various spectroscopic >experiments such as the photoemission, X-ray scattering, and neutron >scattering experiments. > Current: undetermined amount >Justification: > > Requested: 100000 > >A specific reason has been given: >This year our project involves two persons (PI and his post-doc) and we >need more computational time. > >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 - DopingP and Irridates
by Bair, Raymond A. 22 Jan '16

22 Jan '16

Lets approve 175K each for DopingP and Irridates Ray On 1/22/16, 10:57 AM, "Lopez Bezanilla, Alejandro" <albezanilla(a)anl.gov> wrote: >Dear Raymond, > >Thank you for your e-mails. If we could get 175k hours for each account, >we could keep doing our calculation between now and the end of March. > >Thanks indeed for helping. > >Alejandro > >________________________________________ >From: Bair, Raymond A. >Sent: Friday, January 22, 2016 9:40 AM >To: Lopez Bezanilla, Alejandro >Cc: LCRC Allocations Admins >Subject: Re: Project Allocation Request - DopingP > >Dear Alejandro, > >Could you estimate how much more time you now and the end of March? In >March the LCRC Allocations Committee meets to make decisions about the >time allocations for Q3 and Q4 (April-Sept.). So the immediate decision is >between now and March 31st, and the rest of your request can be added to >your original request for Q3 and Q4. > >Regards, > >Ray > >----------------------------------------- >Ray Bair >Argonne National Laboratory >and the University of Chicago > > > > > > > > > >On 1/21/16, 11:48 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: albezanilla (Alejandro Lopez Bezanilla) >> Project: DopingP >> Title: Transport Properties Of Doped Phosphorene Nanoribbons >> Description: The project is progressing based on two types on >>calculations related to both electronic structure and transport >>properties of defected phosphorene structures. A large set of defected >>structures ranging from substitutional doping to vacancies are in >>progress. >>The code used for the first-principles calculations is SIESTA running in >>Blues. Multiples job running on one or two nodes for several days >>(non-stop) are submitted. The performance is good considering the large >>number of atoms in each super-cell. >>The transport calculations are submitted over 16 nodes in Blues for >>several hours each. >>More defects are being modelled currently and approximately 225k hours of >>computational time are estimated to complete the study. We expect to >>finish the study with the next 3 quarters. >> Current: undetermined amount >>Justification: >> >> Requested: 200000 >> >>A specific reason has been given: >>Large P based system composed of hundreds of atoms are being treated with >>DFT tools. A systematic exploration of different types of defects is in >>process of execution and additional computational time is needed for the >>quantum transport calculations. >> >>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

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 22 Jan '16

22 Jan '16

Hello, A change in allocation has been requested: Requester: hyowon (Hyowon Park) Project: dmft_for_oxides Title: DFT+DMFT Calculations of Transition Metal Oxides Description: A DFT+DMFT code consists of a DFT part and a DMFT part. Most of modern DFT codes are efficiently parallelized such that simulations of large supercells can be performed. A DMFT calculation requires an accurate convergence of the quantum Monte Carlo method. The quantum Monte Carlo simulation is scalable almost linearly as the number of cpus. These codes can be compiled using the Intel compilers and MPI which are already installed in LCRC machines. Therefore, the parallel calculation using the clusters in LCRC will be inevitable for performing precise and efficient DFT+DMFT simulations. During FY 2015-16, I will perform DFT+DMFT calculations to study an accurate ground-state description of strongly correlated materials including structure, total energy, and magnetism as well as excited-state properties such as the one-particle spectral function and two-particle magnetic/charge response functions. Specifically, I will study the strong spin-lattice-charge coupling in La1/3Sr2/3FeO3 oxides, the magnetic susceptibility in CePd3 heavy fermions, the spin-state transition in LaCoO3, the metal-insulator transition in rare-earth nickelate thin films/superlattices, and the energetics of UO2 heavy fermions. These DFT+DMFT outputs can be directly compared to various spectroscopic experiments such as the photoemission, X-ray scattering, and neutron scattering experiments. Current: undetermined amount Justification: Requested: 100000 A specific reason has been given: This year our project involves two persons (PI and his post-doc) and we need more computational time. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] Project Allocation Request - NEXTGENOPT
by Bair, Raymond A. 22 Jan '16

22 Jan '16

Lets approve the request for 160K. Ray On 1/21/16, 3:45 PM, "Kim, Kibaek" <kimk(a)anl.gov> wrote: >Dear Ray: > >Here is my estimate: > >1. A research project that is currently using resources will further >require 160000 core-hours in a month. >2. Another research project that will start running computations in >February is expected to require 50000 core-hours. > >So, I would need nearly 160000 core-hours more by the end of March. > >Best, >Kibaek > >> On Jan 21, 2016, at 9:34 AM, Bair, Raymond A. <rbair(a)anl.gov> wrote: >> >> Dear Kibaek Kim, >> >> We need your feedback to proceed. >> >> Regards, >> >> Ray >> >> >> >> On 1/14/16, 4:21 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: >> >>> Dear Kibaek, >>> >>> Could you estimate how much more time you now and the end of March? In >>> March the LCRC Allocations Committee meets to make decisions about the >>> time allocations for Q3 and Q4 (April-Sept.). So the immediate >>>decision >>> is between now and March 31st, and the rest of your request can be >>>added >>> to your original request for Q3 and Q4. >>> >>> Regards, >>> >>> Ray >>> >>> >>> ----------------------------------------- >>> Ray Bair >>> Argonne National Laboratory >>> and the University of Chicago >>> >>> >>> >>> >>> >>> >>> >>> >>> On 1/14/16, 10:00 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: kibaekkim (Kibaek Kim) >>>> Project: NEXTGENOPT >>>> Title: Next generation optimization >>>> Description: Sponsored by a Department of Energy (DOE) Early Career >>>> Award, we are currently developing new algorithms for UQ/SO that have >>>>a >>>> holistic view of the entire data-modeling-optimization process. A >>>> particular example of such algorithms is a new clustering-based >>>> interior-point strategy for stochastic programs in which scenarios are >>>> compressed based on their influence on the first-stage decision, not >>>>on >>>> data. This has shown to lead to drastic compression rates (and >>>>solution >>>> times) of above 80%, which cannot be achieved with data compression >>>> alone (see Figure 1). The compression is done adaptively, at the >>>>linear >>>> algebra level, by looking at the different contributions of the >>>> scenarios on the Schur complement. >>>> >>>> We are also currently developing new interior-point algorithms for >>>> nonconvex structured optimization capable of dealing with complex >>>> multi-scale physics (typically involving large networks of partial >>>> differential equations) and stochastic components. These problems >>>>arise, >>>> for instance, from natural gas inventory (line-pack) management (see >>>> Figure 2). This requires of a redesign of existing algorithms and >>>> modeling environments capable of conveying problem structure down to >>>>the >>>> linear algebra kernels. In addition, hybrid linear algebra kernels >>>>need >>>> to be designed to exploit different structural facets. Problems with >>>> billions of variables are being targeted. >>>> Current: undetermined amount >>>> Justification: >>>> >>>> Requested: 400000 >>>> >>>> A specific reason has been given: >>>> We are running a large number of optimization instances for a power >>>>grid >>>> application. The application analyzes the cost efficiency of the large >>>> power grid system in California with uncertain wind power production. >>>>As >>>> a result, a large scale stochastic optimization problem needs to be >>>> solved with different seasons and different parameter settings, which >>>> requires more core-hours to run such optimization problems. >>>> >>>> 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] Project Allocation Request - DopingP
by Bair, Raymond A. 22 Jan '16

22 Jan '16

Dear Alejandro, Could you estimate how much more time you now and the end of March? In March the LCRC Allocations Committee meets to make decisions about the time allocations for Q3 and Q4 (April-Sept.). So the immediate decision is between now and March 31st, and the rest of your request can be added to your original request for Q3 and Q4. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/21/16, 11:48 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: albezanilla (Alejandro Lopez Bezanilla) > Project: DopingP > Title: Transport Properties Of Doped Phosphorene Nanoribbons > Description: The project is progressing based on two types on >calculations related to both electronic structure and transport >properties of defected phosphorene structures. A large set of defected >structures ranging from substitutional doping to vacancies are in >progress. >The code used for the first-principles calculations is SIESTA running in >Blues. Multiples job running on one or two nodes for several days >(non-stop) are submitted. The performance is good considering the large >number of atoms in each super-cell. >The transport calculations are submitted over 16 nodes in Blues for >several hours each. >More defects are being modelled currently and approximately 225k hours of >computational time are estimated to complete the study. We expect to >finish the study with the next 3 quarters. > Current: undetermined amount >Justification: > > Requested: 200000 > >A specific reason has been given: >Large P based system composed of hundreds of atoms are being treated with >DFT tools. A systematic exploration of different types of defects is in >process of execution and additional computational time is needed for the >quantum transport calculations. > >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 - Irridates
by Bair, Raymond A. 22 Jan '16

22 Jan '16

Dear Alejandro, Could you estimate how much more time you now and the end of March? In March the LCRC Allocations Committee meets to make decisions about the time allocations for Q3 and Q4 (April-Sept.). So the immediate decision is between now and March 31st, and the rest of your request can be added to your original request for Q3 and Q4. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/21/16, 11:42 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: albezanilla (Alejandro Lopez Bezanilla) > Project: Irridates > Title: Metal-insulator phase transitions in irridates > Description: We will use MPI Fortran codes developed at >Coulumbia University at NYC. Based on previous >cluster DMFT calculations for a model tight >binding Hamiltonian we estimate the CPU hours for >a submitted job as 10k: 12 nodes * 16 cores per >node (Blues) running for 48 hours. >The work will be divided in multiple sets >containing external parameters. Considering the >limited resources, we will target parameters to >solve instead of surveying entire phase diagram. >Previous tests using thousands of cores >simultaneously indicate that a better performance >is achieved running jobs on hundreds of cores. >The project is expected to have three members at >most. > Current: undetermined amount >Justification: > > Requested: 350000 > >A specific reason has been given: >Our DMFT calculations uses configuration-interaction-based >exact diagonalization solvers which are very time consuming > >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

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 21 Jan '16

21 Jan '16

Hello, A change in allocation has been requested: Requester: albezanilla (Alejandro Lopez Bezanilla) Project: DopingP Title: Transport Properties Of Doped Phosphorene Nanoribbons Description: The project is progressing based on two types on calculations related to both electronic structure and transport properties of defected phosphorene structures. A large set of defected structures ranging from substitutional doping to vacancies are in progress. The code used for the first-principles calculations is SIESTA running in Blues. Multiples job running on one or two nodes for several days (non-stop) are submitted. The performance is good considering the large number of atoms in each super-cell. The transport calculations are submitted over 16 nodes in Blues for several hours each. More defects are being modelled currently and approximately 225k hours of computational time are estimated to complete the study. We expect to finish the study with the next 3 quarters. Current: undetermined amount Justification: Requested: 200000 A specific reason has been given: Large P based system composed of hundreds of atoms are being treated with DFT tools. A systematic exploration of different types of defects is in process of execution and additional computational time is needed for the quantum transport calculations. This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

[LCRC Accounts] Project Allocation Request
by accounts＠lcrc.anl.gov 21 Jan '16

21 Jan '16

Hello, A change in allocation has been requested: Requester: albezanilla (Alejandro Lopez Bezanilla) Project: Irridates Title: Metal-insulator phase transitions in irridates Description: We will use MPI Fortran codes developed at Coulumbia University at NYC. Based on previous cluster DMFT calculations for a model tight binding Hamiltonian we estimate the CPU hours for a submitted job as 10k: 12 nodes * 16 cores per node (Blues) running for 48 hours. The work will be divided in multiple sets containing external parameters. Considering the limited resources, we will target parameters to solve instead of surveying entire phase diagram. Previous tests using thousands of cores simultaneously indicate that a better performance is achieved running jobs on hundreds of cores. The project is expected to have three members at most. Current: undetermined amount Justification: Requested: 350000 A specific reason has been given: Our DMFT calculations uses configuration-interaction-based exact diagonalization solvers which are very time consuming This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] Project Allocation Request - NEXTGENOPT
by Bair, Raymond A. 14 Jan '16

14 Jan '16

Dear Kibaek, Could you estimate how much more time you now and the end of March? In March the LCRC Allocations Committee meets to make decisions about the time allocations for Q3 and Q4 (April-Sept.). So the immediate decision is between now and March 31st, and the rest of your request can be added to your original request for Q3 and Q4. Regards, Ray ----------------------------------------- Ray Bair Argonne National Laboratory and the University of Chicago On 1/14/16, 10:00 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: kibaekkim (Kibaek Kim) > Project: NEXTGENOPT > Title: Next generation optimization > Description: Sponsored by a Department of Energy (DOE) Early Career Award, we are currently developing new algorithms for UQ/SO that have a holistic view of the entire data-modeling-optimization process. A particular example of such algorithms is a new clustering-based interior-point strategy for stochastic programs in which scenarios are compressed based on their influence on the first-stage decision, not on data. This has shown to lead to drastic compression rates (and solution times) of above 80%, which cannot be achieved with data compression alone (see Figure 1). The compression is done adaptively, at the linear algebra level, by looking at the different contributions of the scenarios on the Schur complement. > >We are also currently developing new interior-point algorithms for nonconvex structured optimization capable of dealing with complex multi-scale physics (typically involving large networks of partial differential equations) and stochastic components. These problems arise, for instance, from natural gas inventory (line-pack) management (see Figure 2). This requires of a redesign of existing algorithms and modeling environments capable of conveying problem structure down to the linear algebra kernels. In addition, hybrid linear algebra kernels need to be designed to exploit different structural facets. Problems with billions of variables are being targeted. > Current: undetermined amount >Justification: > > Requested: 400000 > >A specific reason has been given: >We are running a large number of optimization instances for a power grid application. The application analyzes the cost efficiency of the large power grid system in California with uncertain wind power production. As a result, a large scale stochastic optimization problem needs to be solved with different seasons and different parameter settings, which requires more core-hours to run such optimization problems. > >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