October 2013 - Allocations-admins - mailman3.cels.anl.gov

[LCRC Accounts] Project Request: MatModelCourse
by accounts＠lcrc.anl.gov 30 Oct '13

30 Oct '13

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: John J. Low Applicant's institution: ANL Applicant's division: CLS Project Name: MatModelCourse Project title: Materials Modeling Course Associated funding: None Other Systems: None Science: This allocation will support the Materials Modeling Course by Ron Miller and Elad Tadmor that I will be hosting. Project description: This course will teach the fundamentals of Materials Modeling from the nano to macro scales. A hands on tutorial on fusion will be used to illustrate application of the concepts taught by the course. Project URL: http://www.modelingmaterials.org/short-courses Requested allocation: 10000 Q1: 10000 Q2: 0 Q3: 0 Q4: 0 Justification: 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 Request: disceng
by accounts＠lcrc.anl.gov 29 Oct '13

29 Oct '13

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: Justin M Wozniak Applicant's institution: ANL Applicant's division: MCS Project Name: disceng Project title: Discovery Engines Associated funding: ANL LDRD Other Systems: We are using the remnants of PADS for interactive data analysis and visualization. Science: The science objectives of this project are to improve data management and analysis techniques for X-ray crystallography and improve experiment analysis turnaround time. This will include data movement improvements as well as novel use of simulation-based analysis. Project description: The initial computational goals are limited: 1) Store/archive data as it comes in from the APS 2) Restore data if data loss occurs on PADS 3) Perform initial studies of scalability of novel DIFFEV/Swift application The programming model used is Swift/T, which executes as an MPI program at run time. The data management will be performed with Globus Online. No known scalability issues exist- Swift/T scales to hundreds of thousands of cores. We plan to experiment with realtime data streaming from experiments at the APS to PADS for visualization using LCRC as a persistent store. We will also experiment with various software integration techniques for DIFFEV/Swift. We have no software requirements above Fusion defaults. No previous runs have been performed- this is very preliminary. The data management operations will include little or no calculation. The DIFFEV/Swift work has the potential to utilize up to 100,000 cores. The existing DIFFEV is known to be very efficient. We expect about 10 project members. No single core runs are planned. Project URL: Requested allocation: 8000 Q1: 2000 Q2: 2000 Q3: 2000 Q4: 2000 Justification: 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 Request: MHK
by accounts＠lcrc.anl.gov 24 Oct '13

24 Oct '13

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: Mark Grippo Applicant's institution: ANL Applicant's division: EVS Project Name: MHK Project title: Hydrodynamic model development for Cobscook Bay Associated funding: DOE Other Systems: Science: Hydrology Project description: Create high resolution hydrodynamic model for Cobscook Bay Maine Project URL: Requested allocation: 1088640 Q1: 272160 Q2: 272160 Q3: 272160 Q4: 272160 Justification: Our run (with turbines activated) simulated 6 days of tides in 12 days of computer time, i.e., 12 real time days. That comes to 12 days * 24 hours * 360 cores = 103680 core-hours for a 6 day tidal simulation. This is for our marconi cluster at UMaine. The rate is 130680 core.hours / (6 days*24 hours*360 cores) = 2 core-hours per 1 hour of tides simulated per core. For the Fusion cluster, we have 800 nodes at roughly the same cpu speed and available memory. So, the rate is 103680 core.hours /( 6 days*24 hours*800 cores ) = 0.9 core-hour per 1 hour of tides simulated per core. We expect to do 7 simulations. 5 simulations of experiments (each is 7 days long) 2 simulation of calibration (each is 14 days long) This gives (5 runs * 7 days) +( 2 runs *14 days) = 63 days of tidal simulations = 1512 hours of tidal simulations. That translates to: 0.9 core-hour per 1 hour of tidal simulated per core * 1512 hours of tidal simulation = 1361 core-hours per core. That comes to 1361 core-hours per core *800 cores=1,088,640 core hours. 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

New LCRC report submitted for Dow-scaling
by support＠lcrc.anl.gov 14 Oct '13

14 Oct '13

Full project report at https://accounts.lcrc.anl.gov/reporting/project/2013/?project=Dow-scaling

1 0

Re: [allocations-admins] [LCRC Accounts] Yearly Allocation Request from Concerted-Flows
by Bair, Raymond A. 11 Oct '13

11 Oct '13

Seems fine to me. Ray Sent from my iPhone > On Oct 11, 2013, at 4:47 PM, "John Valdes" <valdes(a)anl.gov> wrote: > > OK to renew this project? Ketan Maheshwari (primary user of this > project) is hoping to run some jobs for an upcoming paper deadline. > They didn't submit a project report for FY13, but the project started > in late May, only a few days before June 1, our stated date for > waiving the report requirement, so I'm OK with waiving the report > requirement for them. > > John > >> On Fri, Oct 11, 2013 at 03:57:25PM -0500, accounts(a)lcrc.anl.gov wrote: >> Hello, >> >> A yearly allocation for the LCRC cluster has been requested with the >> following updated information: >> >> Submitter/PI: Raj Kettimuthu >> Project Name: Concerted-Flows >> Division: MCS >> Project title: Concerted Flows: Infrastructure for Data Intensive Science on High Speed Networks >> Associated funding: DOE ASCR >> Other Systems: >> Science: Our objective is to use model-driven approaches to improve the performance of distributed data and compute intensive applications. There has been a significant increase in the number of scientific applications over the recent years. Meanwhile, a wide variety of computational sites have emerged with shared access to users such that a user often has accesses to multiple sites, for each site the user has a limited chunk of resource allocation. Often times, it is not possible or not optimal to execute all the component tasks of a scientific workflow on a single site. Our goal is to develop optimization heuristics to judiciously schedule individual tasks in the workflow to resources at multiple sites to balance the workload among sites and minimize the overhead in data transfer. >> Project description: Our applications will be loosely coupled - parameter sweep or bag of tasks - often with multiple stages. We have an immediate requirement to evaluate our scheduling strategy for a mineral physics workflow for which we need 65 nodes for several hours to start with. We plan to do several runs of similar types of applications. Expected number of project members is 4. >> Project URL: http://wiki.mcs.anl.gov/concerted-flows/index.php/Main_Page >> Current FY Hours Used: undetermined amount >> >> New FY Requested allocation: 200000 >> Q1: 80000 >> Q2: 40000 >> Q3: 40000 >> Q4: 40000 >> >> Justification: >> >> Thank You, >> >> The LCRC Accounts System >> _______________________________________________ >> allocations-admins mailing list >> allocations-admins(a)lists.lcrc.anl.gov >> https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins > _______________________________________________ > allocations-admins mailing list > allocations-admins(a)lists.lcrc.anl.gov > https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

Re: [allocations-admins] [LCRC Accounts] Yearly Allocation Request from Concerted-Flows
by John Valdes 11 Oct '13

11 Oct '13

OK to renew this project? Ketan Maheshwari (primary user of this project) is hoping to run some jobs for an upcoming paper deadline. They didn't submit a project report for FY13, but the project started in late May, only a few days before June 1, our stated date for waiving the report requirement, so I'm OK with waiving the report requirement for them. John On Fri, Oct 11, 2013 at 03:57:25PM -0500, accounts(a)lcrc.anl.gov wrote: > Hello, > > A yearly allocation for the LCRC cluster has been requested with the > following updated information: > > Submitter/PI: Raj Kettimuthu > Project Name: Concerted-Flows > Division: MCS > Project title: Concerted Flows: Infrastructure for Data Intensive Science on High Speed Networks > Associated funding: DOE ASCR > Other Systems: > Science: Our objective is to use model-driven approaches to improve the performance of distributed data and compute intensive applications. There has been a significant increase in the number of scientific applications over the recent years. Meanwhile, a wide variety of computational sites have emerged with shared access to users such that a user often has accesses to multiple sites, for each site the user has a limited chunk of resource allocation. Often times, it is not possible or not optimal to execute all the component tasks of a scientific workflow on a single site. Our goal is to develop optimization heuristics to judiciously schedule individual tasks in the workflow to resources at multiple sites to balance the workload among sites and minimize the overhead in data transfer. > Project description: Our applications will be loosely coupled - parameter sweep or bag of tasks - often with multiple stages. We have an immediate requirement to evaluate our scheduling strategy for a mineral physics workflow for which we need 65 nodes for several hours to start with. We plan to do several runs of similar types of applications. Expected number of project members is 4. > Project URL: http://wiki.mcs.anl.gov/concerted-flows/index.php/Main_Page > Current FY Hours Used: undetermined amount > > New FY Requested allocation: 200000 > Q1: 80000 > Q2: 40000 > Q3: 40000 > Q4: 40000 > > Justification: > > Thank You, > > The LCRC Accounts System > _______________________________________________ > allocations-admins mailing list > allocations-admins(a)lists.lcrc.anl.gov > https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

[LCRC Accounts] Yearly Allocation Request from Concerted-Flows
by accounts＠lcrc.anl.gov 11 Oct '13

11 Oct '13

Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Raj Kettimuthu Project Name: Concerted-Flows Division: MCS Project title: Concerted Flows: Infrastructure for Data Intensive Science on High Speed Networks Associated funding: DOE ASCR Other Systems: Science: Our objective is to use model-driven approaches to improve the performance of distributed data and compute intensive applications. There has been a significant increase in the number of scientific applications over the recent years. Meanwhile, a wide variety of computational sites have emerged with shared access to users such that a user often has accesses to multiple sites, for each site the user has a limited chunk of resource allocation. Often times, it is not possible or not optimal to execute all the component tasks of a scientific workflow on a single site. Our goal is to develop optimization heuristics to judiciously schedule individual tasks in the workflow to resources at multiple sites to balance the workload among sites and minimize the overhead in data transfer. Project description: Our applications will be loosely coupled - parameter sweep or bag of tasks - often with multiple stages. We have an immediate requirement to evaluate our scheduling strategy for a mineral physics workflow for which we need 65 nodes for several hours to start with. We plan to do several runs of similar types of applications. Expected number of project members is 4. Project URL: http://wiki.mcs.anl.gov/concerted-flows/index.php/Main_Page Current FY Hours Used: undetermined amount New FY Requested allocation: 200000 Q1: 80000 Q2: 40000 Q3: 40000 Q4: 40000 Justification: Thank You, The LCRC Accounts System

1 0

Re: [allocations-admins] [LCRC Accounts] Project Request: Supernova_DES
by Bair, Raymond A. 07 Oct '13

07 Oct '13

Lets grant the 1st half of this one. POC: Pieper Cat/sc: P Phy On 10/6/13 1:26 PM, "accounts(a)lcrc.anl.gov" <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: Stephen Kuhlmann >Applicant's institution: ANL > Applicant's division: HEP > Project Name: Supernova_DES > Project title: Supernova modeling and analysis for Dark Energy >Survey > Associated funding: DOE/Office of Science/High Energy Physics > Other Systems: 32 nodes ANL/HEP division, 50% time > Science: The Dark Energy Survey is a new 5-year DOE >project >to study cosmology, dark energy, and dark matter. It will discover >thousands of new supernova >explosions in distant galaxies. The key to supernova cosmology is the >measurement of the absolute brightness of the explosion. We simulate the >explosion in a parametric way, and we need to model the full 5-year >survey by varying the explosion parameters in a systematic set of >simulations. > Project description: The project is to define the supernova >cosmology uncertainties for the new Dark Energy Survey which will collect >thousands of >supernova explosions. > >This project uses the SNANA simulation package >http://sdssdp62.fnal.gov/sdsssn/SNANA-PUBLIC/ >as well as the SNCosmo analysis package >http://sncosmo.readthedocs.org/en/latest/index.html. >Both packages have been compiled on Fusion. > >The light-curve fits in SNANA, to a single >supernova, takes 30s on a dedicated processor. >We need x10 the expected data statistics in our >simulations to define uncertainties. >The total expected simulation load per quarter is: >50K supernova * 30s * 10 parameters * 20 sets = >83K hours. Each quarter our simulation >will be updated due to improved >knowledge based on the new data now being >recorded. > >This project will involve five members of the >ANL/HEP DES collaboration. > > > > Project URL: http://www.hep.anl.gov/des/ > Requested allocation: 332000 > Q1: 83000 > Q2: 83000 > Q3: 83000 > Q4: 83000 > Justification: > > >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)lists.lcrc.anl.gov >https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins

1 0

[LCRC Accounts] Project Request: Supernova_DES
by accounts＠lcrc.anl.gov 06 Oct '13

06 Oct '13

Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Stephen Kuhlmann Applicant's institution: ANL Applicant's division: HEP Project Name: Supernova_DES Project title: Supernova modeling and analysis for Dark Energy Survey Associated funding: DOE/Office of Science/High Energy Physics Other Systems: 32 nodes ANL/HEP division, 50% time Science: The Dark Energy Survey is a new 5-year DOE project to study cosmology, dark energy, and dark matter. It will discover thousands of new supernova explosions in distant galaxies. The key to supernova cosmology is the measurement of the absolute brightness of the explosion. We simulate the explosion in a parametric way, and we need to model the full 5-year survey by varying the explosion parameters in a systematic set of simulations. Project description: The project is to define the supernova cosmology uncertainties for the new Dark Energy Survey which will collect thousands of supernova explosions. This project uses the SNANA simulation package http://sdssdp62.fnal.gov/sdsssn/SNANA-PUBLIC/ as well as the SNCosmo analysis package http://sncosmo.readthedocs.org/en/latest/index.html. Both packages have been compiled on Fusion. The light-curve fits in SNANA, to a single supernova, takes 30s on a dedicated processor. We need x10 the expected data statistics in our simulations to define uncertainties. The total expected simulation load per quarter is: 50K supernova * 30s * 10 parameters * 20 sets = 83K hours. Each quarter our simulation will be updated due to improved knowledge based on the new data now being recorded. This project will involve five members of the ANL/HEP DES collaboration. Project URL: http://www.hep.anl.gov/des/ Requested allocation: 332000 Q1: 83000 Q2: 83000 Q3: 83000 Q4: 83000 Justification: 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

FW: [lcrc-core] new projects metadata
by Bair, Raymond A. 04 Oct '13

04 Oct '13

Allocations Committee, Here are the Point of Contact assignments for all of the new FY14 LCRC projects. Please let us know if there is a better POC for some project. Note that we created a new major category of projects. "I" projects are Industry Partnerships, and their default system is Fusion, except when Blues is essential for open work. Regards, Ray Project PI POC Prio Cat:Subcat ------- -- --- ---- --------- ALCF_BOEING_BLUES Ramesh Balakrishnan Shashi Aithal 3 I Trans Anharmonic_systems Darya Aleinikava Mike Minkoff 3 P Chem Bio_catalysis Cong Liu Larry Curtiss 3 P Mat Sci Bubble_U-X Zhigang Mei Elia Merzari 3 P Mat Sci CeO2-voids-bubbles Zhigang Mei Elia Merzari 3 P Mat Sci DFT_NiOxHy_on_Pt3Ni Zhenhua Zeng Larry Curtiss 3 P Mat Sci dual-fuel-sprays Zihan Wang Shashi Aithal 3 I trans eletrolyte_redoxflow Lei Cheng Larry Curtiss 1 P Chem FIX SPELLING electrolyte_redoxflow LES-Engine Sibendu Som Shashi Aithal 3 I Trans metadisolve John Low Larry Curtiss 3 P Mat Sci SHARP_ARC Justin Thomas Elia Merzari 1 A Nuc Eng 200,000 core-hours or less: DopedDWNT Alejandro Bezanilla Larry Curtiss 3 P Mat Sci Li-MetalOxide Alper Kinaci Larry Curtiss 3 P Nano Li_trans_LiOH Alper Kinaci Larry Curtiss 3 P Nano Meso Olle Heinonen Larry Curtiss 3 P Mat Sci NE_RANS Justin Thomas Elia Merzari 3 A Nuc Eng PETSc-Foam Shashi Aithal Misun Min 4 P App Math SRF-materials Denise Ford Larry Curtiss 3 P Mat Sci STOandPlus Alejandro Bezanilla Larry Curtiss 3 P Mat VERITAS Benjamin Zitzer Steve Pieper 3 P Phy Renewal projects that are now mrked as Industry Partnerships are: CAT-engine-modeling (Sibendu Som/CAT) Hospital_Microbiome - (Daniel Smith ,Alfred P. Sloan Foundation) LES-sprays (Sibendu - Cummins) biodiesel-combustion (Sibendu Som - Ford) cavitation-modeling (Sibendu Som - GMRD) Categories (Cat) ---------------- S: Strategic Initiative P: Physical Sciences A: Engineering Applications I: Industry Partnerships M: Misc Subcategories (Subcat) ---------------------- App Math: Applied Mathematics Bio: Biology Chem: Chemistry Cli: Climate Modeling Comp Sci: Computer Sciences Eng App: Other Engineering Apps Env Sci: Environmental Science Geo: Geology Grid: Grid Development and Application Mat Sci: Material Sciences Nano: Nanotechnology Nuc Eng: Nuclear Engineering Phy: Physics Soft: Software Development with Application to Jazz Supp: Support Trans: Transportation

1 0