Award Abstract # 1250171
CDS&E: Data Management and Visualization in Petascale Turbulent Combustion Simulation

NSF Org: CBET
Division of Chemical, Bioengineering, Environmental, and Transport Systems
Recipient: UNIVERSITY OF PITTSBURGH - OF THE COMMONWEALTH SYSTEM OF HIGHER EDUCATION
Initial Amendment Date: September 6, 2012
Latest Amendment Date: August 24, 2015
Award Number: 1250171
Award Instrument: Standard Grant
Program Manager: Ron Joslin
rjoslin@nsf.gov
 (703)292-7030
CBET
 Division of Chemical, Bioengineering, Environmental, and Transport Systems
ENG
 Directorate for Engineering
Start Date: September 15, 2012
End Date: August 31, 2017 (Estimated)
Total Intended Award Amount: $500,000.00
Total Awarded Amount to Date: $500,000.00
Funds Obligated to Date: FY 2012 = $500,000.00
History of Investigator:
  • Peyman Givi (Principal Investigator)
    peg10@pitt.edu
  • Panos Chrysanthis (Co-Principal Investigator)
  • Alexandros Labrinidis (Co-Principal Investigator)
  • Server Yilmaz (Co-Principal Investigator)
  • Georgeta-Elisab Marai (Co-Principal Investigator)
Recipient Sponsored Research Office: University of Pittsburgh
4200 FIFTH AVENUE
PITTSBURGH
PA  US  15260-0001
(412)624-7400
Sponsor Congressional District: 12
Primary Place of Performance: University of Pittsburgh
PA  US  15213-2303
Primary Place of Performance
Congressional District:
12
Unique Entity Identifier (UEI): MKAGLD59JRL1
Parent UEI:
NSF Program(s): CDS&E
Primary Program Source: 01001213DB NSF RESEARCH & RELATED ACTIVIT
Program Reference Code(s): 064E, 7433
Program Element Code(s): 808400
Award Agency Code: 4900
Fund Agency Code: 4900
Assistance Listing Number(s): 47.041

ABSTRACT

CBET-1250171
PI: Peyman Givi, Univ. of Pittsburgh

This award provides funding for developments of (1) a robust computational method for petascale simulation
of turbulent combustion, (2) a scalable data management system for such simulations and (3) a systematic
visual analysis of the generated data. The computational method will be based on the filtered density function
(FDF) methodology for large eddy simulation (LES) of complex turbulent flames. The simulation will
be based on a new algorithm, termed ?irregularly portioned Lagrangian Monte Carlo-Finite Difference?
(IPLMCFD) which facilitate FDF simulations on massively parallel, up to petascale platforms. Data management
will be provided by addressing research challenges in management of annotations, management
of workflows and data archiving. Effective data visualization and analysis will be conducted through a
machine learning ?feature-extraction? approach. This work crosses the disciplines of engineering and computer
science and expands the state-of-the-art in high fidelity predictions of turbulent reacting flows. At the
conclusion of the work, an open source LES code will be provided for use by the public.

If successful, the results of this research will have a significant impact in combustion, both in gas-turbine
industry and in government. It is firmly believed that LES will constitute the primary means of predictions
for future design and manufacturing of combustion systems. Having it coupled with robust and versatile data
management and visualization capabilities will be useful for both basic and applied research purposes. Some
of the other broader impacts are through involvement of undergraduate students in research and attracting
them to graduate school, K-12 outreach, involvement of high school students in research, and recruitment
of students from minority and under-represented groups.

PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH

Note:  When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

(Showing: 1 - 10 of 12)
Angen Zheng, Alexandros Labrinidis, Christos Faloutsos "Skew-Resistant Graph Partitioning" 33nd IEEE International Conference on Data Engineering (ICDE 2017) , 2017 , p.151 10.1109/ICDE.2017.62
Angen Zheng, Alexandros Labrinidis, Panos K. Chrysanthis "Planar: Parallel Lightweight Architecture-Aware Adaptive Graph Repartitioning" 32nd IEEE International Conference on Data Engineering (ICDE) , 2016 , p.121 http://dx.doi.org/10.1109/ICDE.2016.7498234
Angen Zheng, Alexandros Labrinidis, Panos K. Chrysanthis, and Jack Lange "ARGO: Architecture-Aware Graph Partitioning" 2016 IEEE International Conference on Big Data , 2016 , p.284 10.1109/BigData.2016.7840614
Angen Zheng, Alexandros Labrinidis, Patrick H. Pisciuneri, Panos K. Chrysanthis, Peyman Givi "PARAGON: Parallel Architecture-Aware Graph Partition Refinement Algorithm" 19th International Conference on Extending Database Technology (EDBT) , 2016 , p.365 http://dx.doi.org/10.5441/002/edbt.2016.34
B. Habtegiorgis, E. Kreft, T. Luciani, G.E. Marai "Visual Encodings for Immersive Visualization of Turbulent Combustion Data" Tapia Conference 2016 , 2016
G.E. Marai, T. Luciani, A. Maries, L. Yilmaz, M. Nik "Visual Descriptors for Dense Tensor Fields in Computational Turbulent Combustion: A Case Study" Journal Imaging Science and Technology , v.1 , 2016
G.E. Marai, T. Luciani, A. Maries, L. Yilmaz, M. Nik "Visual Descriptors for Dense Tensor Fields in Computational Turbulent Combustion: A Case Study" Journal Imaging Science and Technology , 2016
H. V. Jagadish, Johannes Gehrke, Alexandros Labrinidis, Yannis Papakonstantinou, Jignesh M. Patel, Raghu Ramakrishnan, Cyrus Shahabi "Big Data and Its Technical Challenges" Communications of the ACM , v.57 , 2014 , p.86 10.1145/2611567
N. Ansari, P.A. Strakey, G.M. Goldin and P. Givi "Filtered density function simulation of a realisticswirled combustor" Proc. Combust. Inst. , v.35 , 2015 , p.1433
P.H. Pisciuneri, S.L. Yilmaz, P.A. Strakey, and P.\ Givi, "An Irregularly Portioned FDF Simulator" SIAM J. Sci. Comp. , v.35 , 2013 , p.C438-C452
S.L. Yilmaz, N. Ansari, P.H. Pisciuneri, M.B. Nik, C.C. Otis, and P. Givi, "Applied Filtered Density Function" J. App. Fluid Mech. , v.6 , 2013 , p.311
(Showing: 1 - 10 of 12)

PROJECT OUTCOMES REPORT

Disclaimer

This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.

This project brought together experts in turbulent combustion simulation, data management, and data visualization to address the scale challenges of modern combustion flow simulations. Towards this the team considered (a) the modeling point of view (i.e., how to make the existing models scale better), (b) the systems point of view (i.e., how to make the computation more efficient), and (c) the human point of view (i.e., how to make the results more understandable). This project has had a significant impact on the training of the next generation of engineers and computer scientists. It has involved eight graduate students (including three members of underrepresented groups) and five undergraduate students (including two members of underrepresented groups). In particular, one of the students' work was published at the Tapia conference. This work was instrumental in helping the minority student secure two research fellowships, one from the Computing Research Association and one from the Tapia conference. We have demonstrated results from this project in multiple K-12 open houses and outreach events (5-10 per year). This project has created abstractions and hybrid algorithms to address scalability challenges in the visualization of high-density, petascale engineering tensor fields. This research impacts turbulent combustion design and simulation, both in gas-turbine industry and in research. The resulting models and simulations pave the way to efficient energy consumption and pollution control. Some of the technical contributions of this project are: Development of a new computational methodology for turbulent combustion simulation, termed irregularly portioned Lagrangian Monte Carlo/finite difference" (IPLMCFD), that can scale to thousands of processors. Investigating the use of general purpose graphics processing units (GPUs) for parallel database processing in conjunction with multicore central processing units (CPUs). Dynamic load balancing via graph partitioning and repartitioning techniques that consider the heterogeneity of the underlying hardware (including the network infrastructure). Design of scalable visual abstractions and algorithms for identifying, tracking, and mining regions of interest in turbulent combustion tensor data. These abstractions and algorithms have been implemented in a visual prototype tool Development of a prototype virtual reality visualization of turbulent combustion data using UIC's Electronic Visualization Lab cutting-edge CAVE2 immersive environment. This project has had a significant impact on the training of the next generation of engineers and computer scientists. It has involved eight graduate students (including three members of underrepresented groups) and five undergraduate students (including two members of underrepresented groups). In particular, one of the students' work was published at the Tapia conference. This work was instrumental in helping the minority student secure two research fellowships, one from the Computing Research Association and one from the Tapia conference. We have demonstrated results from this project in multiple K-12 open houses and outreach events (5-10 per year). This project has created abstractions and hybrid algorithms to address scalability challenges in the visualization of high-density, petascale engineering tensor fields. This research impacts turbulent combustion design and simulation, both in gas-turbine industry and in research. The resulting models and simulations pave the way to efficient energy consumption and pollution control. Some of the technical contributions of this project are: Development of a new computational methodology for turbulent combustion simulation, termed irregularly portioned Lagrangian Monte Carlo/finite difference" (IPLMCFD), that can scale to thousands of processors. Investigating the use of general purpose graphics processing units (GPUs) for parallel database processing in conjunction with multicore central processing units (CPUs). Dynamic load balancing via graph partitioning and repartitioning techniques that consider the heterogeneity of the underlying hardware (including the network infrastructure). Design of scalable visual abstractions and algorithms for identifying, tracking, and mining regions of interest in turbulent combustion tensor data. These abstractions and algorithms have been implemented in a visual prototype tool Development of a prototype virtual reality visualization of turbulent combustion data using UIC's Electronic Visualization Lab cutting-edge CAVE2 immersive environment.


Last Modified: 12/19/2017
Modified by: Peyman Givi

Please report errors in award information by writing to: awardsearch@nsf.gov.

Print this page

Back to Top of page