Dynamic Constant Reconfiguration for Explicit Finite Difference Option Pricing

Becker, Tobias; Jin, Qiwei; Luk, Wayne; Weston, Stephen

doi:10.1109/reconfig.2011.29

Cited by 9 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work extends our previous work [4] by making use of carefully chosen coefficients for constant multipliers, so as to reduce the amount of required hardware resources per kernel, and reduce the amount of computation required with given result accuracy requirement. We discuss the application of our work to the financial EFD method, though it is applicable to any stencil computation with constant coefficients.…”

Section: Introductionmentioning

confidence: 68%

“…The result is produced according to the error tolerant level used in industry 2 , which requires the difference between the reduced precision result and the double precision result to be smaller than 2E − 4; however, the workflow can be tuned to accommodate arbitrary error tolerant level by adjusting the number format. The 23 bit fixed point number format with 1 bit for integer and 22 bit for fraction is used in our implementations, as it has been used in our previous work [4] and the finite precision error in the result is always below 2E − 4 compared to double precision floating point result for the EFD computation. The FloPoCo library [11] is used to generate dynamic kernel descriptions with different stencil coefficients in VHDL.…”

Section: Resultsmentioning

confidence: 99%

“…In previous work we have described a parallel hardware architecture which makes use of dynamic constant reconfiguration [4]. The work demonstrates the effectiveness of dynamic constant reconfiguration by studying the use of fixed point constant multipliers in a parallel option pricing architecture, as shown in Figure 2.…”

Section: The Finite Difference Modelmentioning

confidence: 94%

“…For example, using a low precision data path, higher performance can be achieved by increasing sampling points in a numerical integration routine [7]. The second step has been addressed by [4] and [3].…”

Section: Optimising Dynamic Constant Reconfigurationmentioning

confidence: 99%

“…Higher performance and energy efficiency can be achieved by developing reconfigurable versions of an application [3]. In addition, dynamically reconfigurable EFD implementations on FPGAs can have a factor of 4.7 performance improvement over a design using a static configuration [4].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Optimising explicit finite difference option pricing for dynamic constant reconfiguration

Jin

Becker

Luk

et al. 2012

22nd International Conference on Field Programmable Logic and Applications (FPL)

Self Cite

View full text Add to dashboard Cite

This paper demonstrates a novel optimisation methodology to adjust stencil based numerical procedures from the algorithm level, so as to reduce not only the amount of hardware resource consumption per kernel but also the amount of computation required to achieve desired result accuracy, when mapping the algorithm to reconfigurable hardware using dynamic constant reconfiguration. As a result, less area is needed to support run-time reconfiguration, and less computational steps are required in the numerical procedure to obtain a result with given error tolerance. We analyse one thousand fixed point implementations on a Virtex-6 XC6V-LX760 FPGA for randomly generated option pricing problems, which are representative of industrial computation. When comparing optimised implementations to the unoptimised ones, the reconfiguration area upper bound is reduced by 22%; the average number of computational steps is reduced by 23%; and the area-computation-time product is reduced by 40%; while the numerical errors of the results are kept below the error tolerant level used in industry.

show abstract

Section: Introductionmentioning

confidence: 68%

Section: Resultsmentioning

confidence: 99%

Section: The Finite Difference Modelmentioning

confidence: 94%

Section: Optimising Dynamic Constant Reconfigurationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Optimising explicit finite difference option pricing for dynamic constant reconfiguration

Jin

Becker

Luk

et al. 2012

22nd International Conference on Field Programmable Logic and Applications (FPL)

Self Cite

View full text Add to dashboard Cite

show abstract

High-Performance Hardware Acceleration of Asset Simulations

Schryver

Marxen

Weithoffer

et al. 2013

High-Performance Computing Using FPGAs

View full text Add to dashboard Cite

EPiCS: Engineering Proprioception in Computing Systems

Becker

Agne

Lewis

et al. 2012

2012 IEEE 15th International Conference on Computational Science and Engineering

Self Cite

View full text Add to dashboard Cite

Abstract-Modern compute systems continue to evolve towards increasingly complex, heterogeneous and distributed architectures. At the same time, functionality and performance are no longer the only aspects when developing applications for such systems, and additional concerns such as flexibility, power efficiency, resource usage, reliability and cost are becoming increasingly important. This does not only raise the question of how to efficiently develop applications for such systems, but also how to cope with dynamic changes in the application behaviour or the system environment.The EPiCS Project aims to address these aspects through exploring self-awareness and self-expression. Self-awareness allows systems and applications to gather and maintain information about their current state and environment, and reason about their behaviour. Self-expression enables systems to adapt their behaviour autonomously to changing conditions. Innovations in EPiCS are based on systematic integration of research in concepts and foundations, customisable hardware/software platforms and operating systems, and self-aware networking and middleware infrastructure. The developed technologies are validated in three application domains: computational finance, distributed smart cameras and interactive mobile media systems.

show abstract

Dynamic Constant Reconfiguration for Explicit Finite Difference Option Pricing

Cited by 9 publications

References 15 publications

Optimising explicit finite difference option pricing for dynamic constant reconfiguration

Optimising explicit finite difference option pricing for dynamic constant reconfiguration

High-Performance Hardware Acceleration of Asset Simulations

EPiCS: Engineering Proprioception in Computing Systems

Contact Info

Product

Resources

About