Istvan Reguly

Research Associate

istvan.reguly@oerc.ox.ac.uk

I am a research associate working on the OP2 and OPS libraries for unstructured nad structured grids and the OPS library for structured grids under Prof. Mike Giles. My interests include high performance computing applied to scientific applications, accelerated on modern heterogeneous architectures such as CPUs, GPUs, or the Xeon Phi. My primery research interest is to raise the level of abstraction of programming for domain scientists, such as physicists, chemists, mathematicians, to enable them to describe their computational problems at a higher level and not worry about the practical implementation issues, such as managing data movement and parallelism.

Publications

I. Z. Reguly, A. Keita, M. B. Giles. Optimizing for performance on Power8. (2015) Centre for Advanced Studies Conference (CASCON 2015), Nov 2-4, Montreal, Canada

I. Z. Reguly, G. R. Mudalige, M. B. Giles. Design and Development of Domain Specific Active Libraries with Proxy Applications. Workshop on Representative Applications (WRAp), IEEE Cluster, Sep 8-11, Chicago, USA 

I. Z. Reguly, E. László, G.R. Mudalige, and M.B. Giles. (2015) Vectorizing unstructured mesh computations for many-core architectures. Concurrency Computat.: Pract. Exper., doi:10.1002/cpe.3621.

I. Z. Reguly, G. R. Mudalige, C. Bertolli, M. B. Giles, A. Betts, P. H. J. Kelly, and D. Radford. Acceleration of a Full-scale Industrial CFD Application with OP2. (2015) IEEE Transactions on Parallel and Distributed Systems PrePrints, doi:10.1109/TPDS.2015.2453972 (online)

S.P.Jammy, N.D.Sandham, G.R. Mudalige, I.Z.Reguly, M.B. Giles. Block Structured Compressible Navier Stokes Solution Using the OPs High-Level abstraction,  27th International Conference on Parallel CFD, May 2015, Montreal, Canada

M.B. Giles, E. Laszlo, I. Reguly, J. Appleyard, J. Demouth, "GPU implementation of finite difference solvers", Proceedings of the Seventh Workshop on High  Performance Computational Finance (WHPCF'14). Held in conjunction with IEEE/ACM Supercomputing 2014(SC'14) (online)

G.R. Mudalige, I.Z. Reguly, M.B. Giles, A.C. Mallinson, W.P. Gaudin, J.A. Herdman, "Performance Analysis of a High-level Abstractions-based Hydrocode on Future Computing Systems".  Proceedings of the 5th international workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS '14). Held in conjunction with IEEE/ACM Supercomputing 2014(SC'14) (online)

I.Z. Reguly, G.R. Mudalige, M.B. Giles, D. Curran and S. McIntosh-Smith, "The OPS Domain Specific Abstraction for Multi-Block Structured Grid Computations". Proceedings of the 4th international workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing(WOLFHPC '14).Held in conjunction with IEEE/ACM Supercomputing 2014(SC'14). (online)

M. B. Giles and I. Z. Reguly. “Trends in high performance computing for engineering calculations”. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences (2014), special issue on Aerodynamics, computers and the environment. 372 20130319; doi: 10.1098/rsta.2013.0319 (online)

I. Z. Reguly, G. R. Mudalige, C. Bertolli, M. B. Giles, A. Betts, P. H. J. Kelly, and D. Radford. “Acceleration of a Full-scale Industrial CFD Application with OP2”, In: submitted to ACM Transactions on Parallel Computing (2014) (arxiv)

I. Z. Reguly, E. László, G. R. Mudalige, and M. B. Giles. “Vectorizing Unstructured Mesh Computations for Many-core Architectures ”. In: Proceedings of the 2014 International Workshop on Programming Models and Applications for Multi- cores and Manycores. PMAM ’14. Orlando, Florida, USA: ACM, 2014. doi: 10. 1145/2560683.2560686. (onlineslides)

I. Z. Reguly and M. Giles. “Finite Element Algorithms and Data Structures on Graphical Processing Units”. In: International Journal of Parallel Programming (2013). issn: 0885-7458. doi: 10.1007/s10766-013-0301-6. (online) Print: April 2015Volume 43Issue 2pp 203-239

L. Rovigatti, P. Šulc, I. Z. Reguly, and F. Romano. “A comparison between parallelisation approaches in molecular dynamics simulations on GPUs”. J. Comput. Chem. 20153618. DOI: 10.1002/jcc.23763 Available at: http://arxiv.org/abs/1401.4350

G. R. Mudalige, M. B. Giles, J. Thiyagalingam, I. Z. Reguly, C. Bertolli, P. H. J. Kelly, and A. E. Trefethen. “Design and initial performance of a high-level unstructured mesh framework on heterogeneous parallel systems”. In: Parallel Computing 39.11 (2013), pp. 669–692. doi: 10.1016/j.parco.2013.09.004. (online)

M. Giles, G. Mudalige, C. Bertolli, P. Kelly, E. László, and I. Z. Reguly. “An Analytical Study of Loop Tiling for a Large-Scale Unstructured Mesh Application”. In: High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion: 2012, pp. 477–482. (online)

M. B. Giles, G. R. Mudalige, B. Spencer, C. Bertolli, and I. Z. Reguly. “Designing OP2 for GPU Architectures”. In: Journal Parallel and Distributed Computing 73.11 (Nov. 2013), pp. 1451–1460. doi: 10.1016/j.jpdc.2012.07.008. (onlinePDF)

I. Z. Reguly and M. B. Giles. “Efficient sparse matrix-vector multiplication on cache-based GPUs.” In: Proceedings of Innovative Parallel Computing (InPar ’12). San Jose, CA. US.: IEEE, May 2012. doi: 10.1109/InPar.2012.6339602. (PDFonline)

G. R. Mudalige, I. Z. Reguly, M. B. Giles, C. Bertolli, and P. H. J. Kelly. “OP2: An Active Library Framework for Solving Unstructured Mesh-based Applications on Multi-Core and Many-Core Architectures.” In: Proceedings of Innovative Parallel Computing (InPar ’12). San Jose, CA. US.: IEEE, May 2012. doi: 10.1109/InPar. 2012.6339594. (online, PDF)

 

 Conferences

I. Z. Reguly and M. B. Giles. “Efficient and scalable sparse matrix-vector multiplication on cache-based GPUs.” In: Sparse Linear Algebra Solvers for High Perfor- mance Computing Workshop. July 8-9, Warwick, UK, 2013. (slides)

I. Z. Reguly, M. B. Giles, G. R. Mudalige, and C. Bertolli. “Finite element methods in OP2 for heterogeneous architectures”. In: European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS 2012). September 10-14, Vienna, Austria, 2012. (slides)

I. Z. Reguly, M. B. Giles, G. R. Mudalige, and C. Bertolli. “OP2: A library for unstructured grid applications on heterogeneous architectures”. In: European Numerical Mathematics and Advanced Applications (ENUMATH 2013). August 26-30, Lausanne, Switzerland, 2013. (slides)

I. Z. Reguly, G. R. Mudalige, C. Bertolli, M. B. Giles, A. Betts, P. H. J. Kelly., and D. Radford. “Acceleration of a Full-scale Industrial CFD Application with OP2”. Poster at GTC (online) and UK Many-Core Developer Conference 2013 (UKMAC’13). December 16, Oxford, UK, 2013. (slides)

Tutorials and Seminars

Design and Implementation of a Scalable Algebraic Multigrid Framework for a Cluster of GPUs (Formerly NVAMG, now AmgX)  - Many-Core Seminar 2012 Michaelmas (slides).

Advanced GPU Programming techniques - Many-Core Seminar 2013 Michaelmas (slides).

Code Generation from High-Level Specifications - Many-Core Seminar 2014 Hilary (slides).

Profiling & Tuning Applications - CUDA training material (slides)

C++ Programming with Thrust - Oxford CUDA Course material (slides)

Activities

Organizing Oxford Many-Core Seminars - This is a series of seminars targeted at researchers at the university across different disciplines (maths, CS, physics, chemistry) who are interested in using high performance computing to solve their research challenges. Slides from past events are available here.

Reviewer: Parallel Computing (2014), Computers & Mathematics with Applications (2013, 2014), ACM Transactions on Mathematical Software (2014)

Program committee: 5th UK Many-Core Conference (2013),  HiStencils 2014 1st International Workshop on High-Performance Stencil Computations (2013)