uu.seUppsala University Publications
Change search
Link to record
Permanent link

Direct link
BETA
Lukarski, Dimitar
Publications (10 of 11) Show all publications
Trost, N., Jimenez, J., Lukarski, D. & Sanchez, V. (2016). Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION. In: Caruge, D Calvin, C Diop, CM Malvagi, F Trama, JC (Ed.), SNA + MC 2013 - Joint International Conference On Supercomputing In Nuclear Applications + Monte Carlo: . Paper presented at Joint 8th International Conference on Supercomputing in Nuclear Applications (SNA) / 4th Monte Carlo Meeting (MC), OCT 27-31, 2013, Paris, FRANCE. , Article ID UNSP 04108.
Open this publication in new window or tab >>Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION
2016 (English)In: SNA + MC 2013 - Joint International Conference On Supercomputing In Nuclear Applications + Monte Carlo / [ed] Caruge, D Calvin, C Diop, CM Malvagi, F Trama, JC, 2016, article id UNSP 04108Conference paper, Published paper (Refereed)
Abstract [en]

COBAYA3 is a multi-physics system of codes which includes two 3D multi-group neutron diffusion codes, ANDES and COBAYA3-PBP, coupled with COBRA-TF, COBRA-IIIc and SUBCHANFLOW sub-channel thermal-hydraulic codes, for the simulation of LWR core transients. The 3D multi-group neutron diffusion equations are expressed in terms of a sparse linear system which can be solved using different iterative Krylov subspace solvers. The mathematical SPARSKIT library has been used for these purposes as it implements among others, external GMRES, PGMRES and BiCGStab solvers. Multi-core CPUs and graphical processing units (GPUs) provide high performance capabilities which are able to accelerate many scientific computations. To take advantage of these new hardware features in daily use computer codes, the integration of the PARALUTION library to solve sparse systems of linear equations is a good choice. It features several types of iterative solvers and preconditioners which can run on both multi-core CPUs and GPU devices without any modification from the interface point of view. This feature is due to the great portability obtained by the modular and flexible design of the library. By exploring this technology, namely the implementation of the PARALUTION library in COBAYA3, we are able to decrease the solution time of the sparse linear systems by a factor 5.15x on GPU and 2.56x on multi-core CPU using standard hardware. These obtained speedup factors in addition to the implementation details are discussed in this paper.

Keywords
COBAYA3, diffusion approximation, PARALUTION, acceleration, parallelization, multi-core, GPUs
National Category
Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-346203 (URN)10.1051/sname/201404108 (DOI)000408930200137 ()
Conference
Joint 8th International Conference on Supercomputing in Nuclear Applications (SNA) / 4th Monte Carlo Meeting (MC), OCT 27-31, 2013, Paris, FRANCE
Available from: 2018-03-15 Created: 2018-03-15 Last updated: 2018-03-15Bibliographically approved
Gupta, R., Lukarski, D., van Gijzen, M. B. & Vuik, C. (2016). Evaluation of the deflated preconditioned CG method to solve bubbly and porous media flow problems on GPU and CPU. International Journal for Numerical Methods in Fluids, 80, 666-683
Open this publication in new window or tab >>Evaluation of the deflated preconditioned CG method to solve bubbly and porous media flow problems on GPU and CPU
2016 (English)In: International Journal for Numerical Methods in Fluids, ISSN 0271-2091, E-ISSN 1097-0363, Vol. 80, p. 666-683Article in journal (Refereed) Published
National Category
Computational Mathematics Computer Sciences
Identifiers
urn:nbn:se:uu:diva-283752 (URN)10.1002/fld.4170 (DOI)000372270600003 ()
Projects
UPMARCeSSENCE
Available from: 2015-09-24 Created: 2016-04-14 Last updated: 2018-01-10Bibliographically approved
Engblom, S. & Lukarski, D. (2016). Fast Matlab compatible sparse assembly on multicore computers. Parallel Computing, 56, 1-17
Open this publication in new window or tab >>Fast Matlab compatible sparse assembly on multicore computers
2016 (English)In: Parallel Computing, ISSN 0167-8191, E-ISSN 1872-7336, Vol. 56, p. 1-17Article in journal (Refereed) Published
National Category
Computer Sciences Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-230821 (URN)10.1016/j.parco.2016.04.001 (DOI)000378964300001 ()
Projects
UPMARCeSSENCE
Available from: 2016-04-06 Created: 2014-08-29 Last updated: 2018-11-12Bibliographically approved
Trost, N., Jiménez, J., Lukarski, D. & Sanchez, V. (2015). Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION. Annals of Nuclear Energy, 82, 252-259
Open this publication in new window or tab >>Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION
2015 (English)In: Annals of Nuclear Energy, ISSN 0306-4549, E-ISSN 1873-2100, Vol. 82, p. 252-259Article in journal (Refereed) Published
National Category
Software Engineering
Identifiers
urn:nbn:se:uu:diva-260598 (URN)10.1016/j.anucene.2014.08.005 (DOI)000358097000030 ()
Projects
UPMARCeSSENCE
Available from: 2014-08-23 Created: 2015-08-21 Last updated: 2018-01-11Bibliographically approved
Anzt, H., Lukarski, D., Tomov, S. & Dongarra, J. (2015). Self-adaptive multiprecision preconditioners on multicore and manycore architectures. In: High Performance Computing for Computational Science – VECPAR 2014: . Paper presented at VECPAR 2014 (pp. 115-123). Springer
Open this publication in new window or tab >>Self-adaptive multiprecision preconditioners on multicore and manycore architectures
2015 (English)In: High Performance Computing for Computational Science – VECPAR 2014, Springer, 2015, p. 115-123Conference paper, Published paper (Refereed)
Place, publisher, year, edition, pages
Springer, 2015
Series
Lecture Notes in Computer Science ; 8969
National Category
Computer Sciences Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-266070 (URN)10.1007/978-3-319-17353-5_10 (DOI)000361760300010 ()978-3-319-17352-8 (ISBN)
Conference
VECPAR 2014
Projects
UPMARCeSSENCE
Available from: 2015-04-18 Created: 2015-11-05 Last updated: 2018-01-10Bibliographically approved
Dorostkar, A., Lukarski, D., Lund, B., Neytcheva, M., Notay, Y. & Schmidt, P. (2014). CPU and GPU performance of large scale numerical simulations in Geophysics. In: Euro-Par 2014: Parallel Processing Workshops, Part I. Paper presented at 1st Workshop on Applications of Parallel Computation in Industry and Engineering (pp. 12-23). Springer
Open this publication in new window or tab >>CPU and GPU performance of large scale numerical simulations in Geophysics
Show others...
2014 (English)In: Euro-Par 2014: Parallel Processing Workshops, Part I, Springer, 2014, p. 12-23Conference paper, Published paper (Refereed)
Place, publisher, year, edition, pages
Springer, 2014
Series
Lecture Notes in Computer Science ; 8805
National Category
Computer Sciences Geophysics
Identifiers
urn:nbn:se:uu:diva-238254 (URN)10.1007/978-3-319-14325-5_2 (DOI)000354783500002 ()978-3-319-14324-8 (ISBN)
Conference
1st Workshop on Applications of Parallel Computation in Industry and Engineering
Projects
UPMARCeSSENCE
Available from: 2014-12-10 Created: 2014-12-10 Last updated: 2018-01-11Bibliographically approved
Lukarski, D. & Neytcheva, M. (2014). On the impact of the heterogeneous multicore and many-core platforms on iterative solution methods and preconditioning techniques. In: High-Performance Computing on Complex Environments: (pp. 13-32). Hoboken, NJ: Wiley-Blackwell
Open this publication in new window or tab >>On the impact of the heterogeneous multicore and many-core platforms on iterative solution methods and preconditioning techniques
2014 (English)In: High-Performance Computing on Complex Environments, Hoboken, NJ: Wiley-Blackwell, 2014, p. 13-32Chapter in book (Refereed)
Place, publisher, year, edition, pages
Hoboken, NJ: Wiley-Blackwell, 2014
National Category
Computer Sciences Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-224645 (URN)10.1002/9781118711897.ch2 (DOI)978-1-118-71205-4 (ISBN)
Projects
UPMARCeSSENCE
Available from: 2014-04-18 Created: 2014-05-15 Last updated: 2018-01-11Bibliographically approved
Dorostkar, A., Lukarski, D., Lund, B., Neytcheva, M., Notay, Y. & Schmidt, P. (2014). Parallel performance study of block-preconditioned iterative methods on multicore computer systems.
Open this publication in new window or tab >>Parallel performance study of block-preconditioned iterative methods on multicore computer systems
Show others...
2014 (English)Report (Other academic)
Series
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2014-007
National Category
Computer Sciences Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-223795 (URN)
Projects
eSSENCEUPMARC
Available from: 2014-03-19 Created: 2014-04-25 Last updated: 2018-01-11Bibliographically approved
Lukarski, D. & Skoglund, T. (2013). A priori power estimation of linear solvers on multi-core processors.
Open this publication in new window or tab >>A priori power estimation of linear solvers on multi-core processors
2013 (English)Report (Other academic)
Series
Technical report / Department of Information Technology, Uppsala University, ISSN 1404-3203 ; 2013-020
National Category
Computer Sciences Computational Mathematics
Identifiers
urn:nbn:se:uu:diva-208505 (URN)
Projects
UPMARCeSSENCE
Available from: 2013-09-30 Created: 2013-10-01 Last updated: 2018-01-11Bibliographically approved
Trost, N., Jiménez, J., Lukarski, D. & Sanchez, V. (2013). Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION. In: Proc. 2nd Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo: . Paper presented at 2nd Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo. Paris, France: La Société Française d'Energie Nucléaire
Open this publication in new window or tab >>Accelerating COBAYA3 on multi-core CPU and GPU systems using PARALUTION
2013 (English)In: Proc. 2nd Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo, Paris, France: La Société Française d'Energie Nucléaire , 2013Conference paper, Published paper (Other academic)
Place, publisher, year, edition, pages
Paris, France: La Société Française d'Energie Nucléaire, 2013
National Category
Software Engineering
Identifiers
urn:nbn:se:uu:diva-212959 (URN)
Conference
2nd Joint International Conference on Supercomputing in Nuclear Applications and Monte Carlo
Projects
UPMARCeSSENCE
Available from: 2013-10-28 Created: 2013-12-16 Last updated: 2018-01-11Bibliographically approved
Organisations

Search in DiVA

Show all publications