Standard
Acceleration of lattice models for pricing portfolios of fixed-income derivatives. / Pawlak, Wojciech Michal; Hlava, Marek; Metaksov, Martin; Oancea, Cosmin Eugen.
ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021. ed. / Tze Meng Low; Jeremy Gibbons. Association for Computing Machinery, Inc., 2021. p. 27-38 3464309.
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
Pawlak, WM, Hlava, M, Metaksov, M
& Oancea, CE 2021,
Acceleration of lattice models for pricing portfolios of fixed-income derivatives. in TM Low & J Gibbons (eds),
ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021., 3464309, Association for Computing Machinery, Inc., pp. 27-38, 7th ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming, ARRAY 2021, held in association with PLDI 2021, Virtual, Online, Canada,
21/06/2021.
https://doi.org/10.1145/3460944.3464309
APA
Pawlak, W. M., Hlava, M., Metaksov, M.
, & Oancea, C. E. (2021).
Acceleration of lattice models for pricing portfolios of fixed-income derivatives. In T. M. Low, & J. Gibbons (Eds.),
ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021 (pp. 27-38). [3464309] Association for Computing Machinery, Inc..
https://doi.org/10.1145/3460944.3464309
Vancouver
Pawlak WM, Hlava M, Metaksov M
, Oancea CE.
Acceleration of lattice models for pricing portfolios of fixed-income derivatives. In Low TM, Gibbons J, editors, ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021. Association for Computing Machinery, Inc. 2021. p. 27-38. 3464309
https://doi.org/10.1145/3460944.3464309
Author
Pawlak, Wojciech Michal ; Hlava, Marek ; Metaksov, Martin ; Oancea, Cosmin Eugen. / Acceleration of lattice models for pricing portfolios of fixed-income derivatives. ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021. editor / Tze Meng Low ; Jeremy Gibbons. Association for Computing Machinery, Inc., 2021. pp. 27-38
Bibtex
@inproceedings{dd97c455e88e4fee9640b4864b43e36d,
title = "Acceleration of lattice models for pricing portfolios of fixed-income derivatives",
abstract = "This paper reports on the acceleration of a standard, lattice-based numerical algorithm that is widely used in finance for pricing a class of fixed-income vanilla derivatives. We start with a high-level algorithmic specification, exhibiting irregular nested parallelism, which is challenging to map efficiently to GPU hardware. From it we systematically derive and optimize two CUDA implementations, which utilize only the outer or all levels of parallelism, respectively. A detailed evaluation demonstrates (i) the high impact of the proposed optimizations, (ii) the complementary strength and weaknesses of the two GPU versions, and that (iii) they are on average 2.4× faster than our well-tuned CPU-parallel implementation (OpenMP+AVX2) running on 104 hardware threads, and by 3-to-4 order of magnitude faster than an OpenMP-parallel implementation using the popular QuantLib library. ",
keywords = "Compilers, Computational Finance, Derivative Pricing, GPGPU (Parallel) Programming",
author = "Pawlak, {Wojciech Michal} and Marek Hlava and Martin Metaksov and Oancea, {Cosmin Eugen}",
note = "Publisher Copyright: {\textcopyright} 2021 ACM.; 7th ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming, ARRAY 2021, held in association with PLDI 2021 ; Conference date: 21-06-2021",
year = "2021",
doi = "10.1145/3460944.3464309",
language = "English",
pages = "27--38",
editor = "Low, {Tze Meng} and Jeremy Gibbons",
booktitle = "ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021",
publisher = "Association for Computing Machinery, Inc.",
}
RIS
TY - GEN
T1 - Acceleration of lattice models for pricing portfolios of fixed-income derivatives
AU - Pawlak, Wojciech Michal
AU - Hlava, Marek
AU - Metaksov, Martin
AU - Oancea, Cosmin Eugen
N1 - Publisher Copyright:
© 2021 ACM.
PY - 2021
Y1 - 2021
N2 - This paper reports on the acceleration of a standard, lattice-based numerical algorithm that is widely used in finance for pricing a class of fixed-income vanilla derivatives. We start with a high-level algorithmic specification, exhibiting irregular nested parallelism, which is challenging to map efficiently to GPU hardware. From it we systematically derive and optimize two CUDA implementations, which utilize only the outer or all levels of parallelism, respectively. A detailed evaluation demonstrates (i) the high impact of the proposed optimizations, (ii) the complementary strength and weaknesses of the two GPU versions, and that (iii) they are on average 2.4× faster than our well-tuned CPU-parallel implementation (OpenMP+AVX2) running on 104 hardware threads, and by 3-to-4 order of magnitude faster than an OpenMP-parallel implementation using the popular QuantLib library.
AB - This paper reports on the acceleration of a standard, lattice-based numerical algorithm that is widely used in finance for pricing a class of fixed-income vanilla derivatives. We start with a high-level algorithmic specification, exhibiting irregular nested parallelism, which is challenging to map efficiently to GPU hardware. From it we systematically derive and optimize two CUDA implementations, which utilize only the outer or all levels of parallelism, respectively. A detailed evaluation demonstrates (i) the high impact of the proposed optimizations, (ii) the complementary strength and weaknesses of the two GPU versions, and that (iii) they are on average 2.4× faster than our well-tuned CPU-parallel implementation (OpenMP+AVX2) running on 104 hardware threads, and by 3-to-4 order of magnitude faster than an OpenMP-parallel implementation using the popular QuantLib library.
KW - Compilers
KW - Computational Finance
KW - Derivative Pricing
KW - GPGPU (Parallel) Programming
U2 - 10.1145/3460944.3464309
DO - 10.1145/3460944.3464309
M3 - Article in proceedings
AN - SCOPUS:85109009544
SP - 27
EP - 38
BT - ARRAY 2021 - Proceedings of the 7th ACM SIGPLAN International Workshop on Libraries, Languages and Compilers for Array Programming, co-located with PLDI 2021
A2 - Low, Tze Meng
A2 - Gibbons, Jeremy
PB - Association for Computing Machinery, Inc.
T2 - 7th ACM SIGPLAN International Workshop on Libraries, Languages, and Compilers for Array Programming, ARRAY 2021, held in association with PLDI 2021
Y2 - 21 June 2021
ER -