Multi-agent Pre-trade Analysis Acceleration in FPGA

Eduardo Gerlein, TM McGinnity, Ammar Belatreche, SA Coleman, Yuhua Li

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Electronic trading in global markets andexchanges requires sophisticated communication and datamanagement systems. Novel computational infrastructures andtrading strategies are required to support the massive amountof incoming streaming data, where the main problem is inlatency management. Multi-agent Systems have been recognizedas a promising solution to address complex problems in manyareas such as biology, social sciences and financial markets andmay provide powerful and flexible solutions for implementingtrading engines. In addition, reconfigurable hardware based onField Programmable Gate Arrays (FPGAs) offers manyimportant performance benefits over software implementations,such as reducing decision making latency and high-throughputdata processing. Robust and scalable trading engines can bedeveloped by leveraging the benefits of reconfigurable FPGAplatforms. This paper presents a multi-agent architecture inreconfigurable hardware for financial applications and theimplementation of a trading engine for pre-trade analysis as avalidation scenario. Performance results show that calculationof technical indicators and trading strategy evaluation togenerate trading signals with a latency of 550 ns is achievable.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages262-269
Number of pages8
Publication statusPublished - 27 Mar 2014
EventIEEE Computational Intelligence for Financial Engineering and Economics - Canary Wharf, London
Duration: 27 Mar 2014 → …

Conference

ConferenceIEEE Computational Intelligence for Financial Engineering and Economics
Period27/03/14 → …

Fingerprint

Field programmable gate arrays (FPGA)
Engines
Reconfigurable hardware
Social sciences
Multi agent systems
Decision making
Hardware
Communication
Processing
Electronic trading
Financial markets

Cite this

Gerlein, E., McGinnity, TM., Belatreche, A., Coleman, SA., & Li, Y. (2014). Multi-agent Pre-trade Analysis Acceleration in FPGA. In Unknown Host Publication (pp. 262-269)
Gerlein, Eduardo ; McGinnity, TM ; Belatreche, Ammar ; Coleman, SA ; Li, Yuhua. / Multi-agent Pre-trade Analysis Acceleration in FPGA. Unknown Host Publication. 2014. pp. 262-269
@inproceedings{81ebc235648c45f38ddb02e65c06ff73,
title = "Multi-agent Pre-trade Analysis Acceleration in FPGA",
abstract = "Electronic trading in global markets andexchanges requires sophisticated communication and datamanagement systems. Novel computational infrastructures andtrading strategies are required to support the massive amountof incoming streaming data, where the main problem is inlatency management. Multi-agent Systems have been recognizedas a promising solution to address complex problems in manyareas such as biology, social sciences and financial markets andmay provide powerful and flexible solutions for implementingtrading engines. In addition, reconfigurable hardware based onField Programmable Gate Arrays (FPGAs) offers manyimportant performance benefits over software implementations,such as reducing decision making latency and high-throughputdata processing. Robust and scalable trading engines can bedeveloped by leveraging the benefits of reconfigurable FPGAplatforms. This paper presents a multi-agent architecture inreconfigurable hardware for financial applications and theimplementation of a trading engine for pre-trade analysis as avalidation scenario. Performance results show that calculationof technical indicators and trading strategy evaluation togenerate trading signals with a latency of 550 ns is achievable.",
author = "Eduardo Gerlein and TM McGinnity and Ammar Belatreche and SA Coleman and Yuhua Li",
year = "2014",
month = "3",
day = "27",
language = "English",
pages = "262--269",
booktitle = "Unknown Host Publication",

}

Gerlein, E, McGinnity, TM, Belatreche, A, Coleman, SA & Li, Y 2014, Multi-agent Pre-trade Analysis Acceleration in FPGA. in Unknown Host Publication. pp. 262-269, IEEE Computational Intelligence for Financial Engineering and Economics, 27/03/14.

Multi-agent Pre-trade Analysis Acceleration in FPGA. / Gerlein, Eduardo; McGinnity, TM; Belatreche, Ammar; Coleman, SA; Li, Yuhua.

Unknown Host Publication. 2014. p. 262-269.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Multi-agent Pre-trade Analysis Acceleration in FPGA

AU - Gerlein, Eduardo

AU - McGinnity, TM

AU - Belatreche, Ammar

AU - Coleman, SA

AU - Li, Yuhua

PY - 2014/3/27

Y1 - 2014/3/27

N2 - Electronic trading in global markets andexchanges requires sophisticated communication and datamanagement systems. Novel computational infrastructures andtrading strategies are required to support the massive amountof incoming streaming data, where the main problem is inlatency management. Multi-agent Systems have been recognizedas a promising solution to address complex problems in manyareas such as biology, social sciences and financial markets andmay provide powerful and flexible solutions for implementingtrading engines. In addition, reconfigurable hardware based onField Programmable Gate Arrays (FPGAs) offers manyimportant performance benefits over software implementations,such as reducing decision making latency and high-throughputdata processing. Robust and scalable trading engines can bedeveloped by leveraging the benefits of reconfigurable FPGAplatforms. This paper presents a multi-agent architecture inreconfigurable hardware for financial applications and theimplementation of a trading engine for pre-trade analysis as avalidation scenario. Performance results show that calculationof technical indicators and trading strategy evaluation togenerate trading signals with a latency of 550 ns is achievable.

AB - Electronic trading in global markets andexchanges requires sophisticated communication and datamanagement systems. Novel computational infrastructures andtrading strategies are required to support the massive amountof incoming streaming data, where the main problem is inlatency management. Multi-agent Systems have been recognizedas a promising solution to address complex problems in manyareas such as biology, social sciences and financial markets andmay provide powerful and flexible solutions for implementingtrading engines. In addition, reconfigurable hardware based onField Programmable Gate Arrays (FPGAs) offers manyimportant performance benefits over software implementations,such as reducing decision making latency and high-throughputdata processing. Robust and scalable trading engines can bedeveloped by leveraging the benefits of reconfigurable FPGAplatforms. This paper presents a multi-agent architecture inreconfigurable hardware for financial applications and theimplementation of a trading engine for pre-trade analysis as avalidation scenario. Performance results show that calculationof technical indicators and trading strategy evaluation togenerate trading signals with a latency of 550 ns is achievable.

M3 - Conference contribution

SP - 262

EP - 269

BT - Unknown Host Publication

ER -

Gerlein E, McGinnity TM, Belatreche A, Coleman SA, Li Y. Multi-agent Pre-trade Analysis Acceleration in FPGA. In Unknown Host Publication. 2014. p. 262-269