Evaluating Versal AI Engines for option price discovery in market risk analysis

Mark Klaisoongnoen; Nick Brown; Tim Dykes; Jessica Jones; Utz-Uwe Haus

doi:10.1145/3626202.3637578

Evaluating Versal AI Engines for option price discovery in market risk analysis

Mark Klaisoongnoen^*, Nick Brown, Tim Dykes, Jessica Jones, Utz-Uwe Haus

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract / Description of output

Whilst Field-Programmable Gate Arrays (FPGAs) have been popular in accelerating high-frequency financial workload for many years, their application in quantitative finance, the utilisation of mathematical models to analyse financial markets and securities, is less mature. Nevertheless, recent work has demonstrated the benefits that FPGAs can deliver to quantitative workloads, and in this paper, we study whether the Versal ACAP and its AI Engines (AIEs) can also deliver improved performance. We focus specifically on the industry standard Strategic Technology Analysis Center's (STAC) derivatives risk analysis benchmark STAC-A2. Porting a purely FPGA-based accelerator STAC-A2 inspired market risk (SIMR) benchmark to the Versal ACAP device by combining Programmable Logic (PL) and AIEs, we explore the development approach and techniques, before comparing performance across PL and AIEs. Ultimately, we found that our AIE approach is slower than a highly optimised existing PL-only version due to limits on both the AIE and PL that we explore and describe.

Original language	English
Title of host publication	Proceedings of the 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays
Publisher	Association for Computing Machinery
Pages	176–182
DOIs	https://doi.org/10.1145/3626202.3637578
Publication status	Published - 15 Mar 2024
Event	The 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays - Monterey, United States Duration: 3 Mar 2024 → 5 Mar 2024 https://www.isfpga.org/program/

Conference

Conference	The 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays
Abbreviated title	ISFPGA
Country/Territory	United States
City	Monterey
Period	3/03/24 → 5/03/24
Internet address	https://www.isfpga.org/program/

Keywords / Materials (for Non-textual outputs)

ai engines
cgras
fpgas
option price discovery
reconfigurable architectures
simr

Access to Document

10.1145/3626202.3637578Licence: All Rights Reserved

ISFPGA24_STAC_A2Accepted author manuscript, 646 KBLicence: Creative Commons: Attribution (CC-BY)

ExCALIBUR HES Exploring Coarse Grained Reconfigurable Architectures (CGRAs)
Brown, N. & Jackson, A.
1/09/22 → 31/05/23
Project: Research

Cite this

@inproceedings{d3792f0bf8d840948c73b25e947e194c,

title = "Evaluating Versal AI Engines for option price discovery in market risk analysis",

abstract = "Whilst Field-Programmable Gate Arrays (FPGAs) have been popular in accelerating high-frequency financial workload for many years, their application in quantitative finance, the utilisation of mathematical models to analyse financial markets and securities, is less mature. Nevertheless, recent work has demonstrated the benefits that FPGAs can deliver to quantitative workloads, and in this paper, we study whether the Versal ACAP and its AI Engines (AIEs) can also deliver improved performance. We focus specifically on the industry standard Strategic Technology Analysis Center's (STAC) derivatives risk analysis benchmark STAC-A2. Porting a purely FPGA-based accelerator STAC-A2 inspired market risk (SIMR) benchmark to the Versal ACAP device by combining Programmable Logic (PL) and AIEs, we explore the development approach and techniques, before comparing performance across PL and AIEs. Ultimately, we found that our AIE approach is slower than a highly optimised existing PL-only version due to limits on both the AIE and PL that we explore and describe.",

keywords = "ai engines, cgras, fpgas, option price discovery, reconfigurable architectures, simr",

author = "Mark Klaisoongnoen and Nick Brown and Tim Dykes and Jessica Jones and Utz-Uwe Haus",

year = "2024",

month = mar,

day = "15",

doi = "10.1145/3626202.3637578",

language = "English",

pages = "176–182",

booktitle = "Proceedings of the 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays",

publisher = "Association for Computing Machinery",

note = "The 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, ISFPGA ; Conference date: 03-03-2024 Through 05-03-2024",

url = "https://www.isfpga.org/program/",

}

Klaisoongnoen, M, Brown, N, Dykes, T, Jones, J & Haus, U-U 2024, Evaluating Versal AI Engines for option price discovery in market risk analysis. in Proceedings of the 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. Association for Computing Machinery, pp. 176–182, The 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, Monterey, California, United States, 3/03/24. https://doi.org/10.1145/3626202.3637578

Evaluating Versal AI Engines for option price discovery in market risk analysis. / Klaisoongnoen, Mark; Brown, Nick; Dykes, Tim et al.
Proceedings of the 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays. Association for Computing Machinery, 2024. p. 176–182.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Evaluating Versal AI Engines for option price discovery in market risk analysis

AU - Klaisoongnoen, Mark

AU - Brown, Nick

AU - Dykes, Tim

AU - Jones, Jessica

AU - Haus, Utz-Uwe

PY - 2024/3/15

Y1 - 2024/3/15

N2 - Whilst Field-Programmable Gate Arrays (FPGAs) have been popular in accelerating high-frequency financial workload for many years, their application in quantitative finance, the utilisation of mathematical models to analyse financial markets and securities, is less mature. Nevertheless, recent work has demonstrated the benefits that FPGAs can deliver to quantitative workloads, and in this paper, we study whether the Versal ACAP and its AI Engines (AIEs) can also deliver improved performance. We focus specifically on the industry standard Strategic Technology Analysis Center's (STAC) derivatives risk analysis benchmark STAC-A2. Porting a purely FPGA-based accelerator STAC-A2 inspired market risk (SIMR) benchmark to the Versal ACAP device by combining Programmable Logic (PL) and AIEs, we explore the development approach and techniques, before comparing performance across PL and AIEs. Ultimately, we found that our AIE approach is slower than a highly optimised existing PL-only version due to limits on both the AIE and PL that we explore and describe.

AB - Whilst Field-Programmable Gate Arrays (FPGAs) have been popular in accelerating high-frequency financial workload for many years, their application in quantitative finance, the utilisation of mathematical models to analyse financial markets and securities, is less mature. Nevertheless, recent work has demonstrated the benefits that FPGAs can deliver to quantitative workloads, and in this paper, we study whether the Versal ACAP and its AI Engines (AIEs) can also deliver improved performance. We focus specifically on the industry standard Strategic Technology Analysis Center's (STAC) derivatives risk analysis benchmark STAC-A2. Porting a purely FPGA-based accelerator STAC-A2 inspired market risk (SIMR) benchmark to the Versal ACAP device by combining Programmable Logic (PL) and AIEs, we explore the development approach and techniques, before comparing performance across PL and AIEs. Ultimately, we found that our AIE approach is slower than a highly optimised existing PL-only version due to limits on both the AIE and PL that we explore and describe.

KW - ai engines

KW - cgras

KW - fpgas

KW - option price discovery

KW - reconfigurable architectures

KW - simr

U2 - 10.1145/3626202.3637578

DO - 10.1145/3626202.3637578

M3 - Conference contribution

SP - 176

EP - 182

BT - Proceedings of the 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays

PB - Association for Computing Machinery

T2 - The 32nd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays

Y2 - 3 March 2024 through 5 March 2024

ER -

Evaluating Versal AI Engines for option price discovery in market risk analysis

Abstract / Description of output

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Projects

ExCALIBUR HES Exploring Coarse Grained Reconfigurable Architectures (CGRAs)

Cite this