Finding the right level of abstraction for minimizing operational expenditure

Authors:
Oskar Mencer;Erik Vynckier;James Spooner;Stephen Girdlestone;Oliver Charlesworth
Affiliations:
Maxeler Technologies, London, United Kingdom;Scottish Widows Investment Partnership, Edinburgh, Scotland Uk;Maxeler Technologies, London, United Kingdom;Maxeler Technologies, London, United Kingdom;Maxeler Technologies, London, United Kingdom
Venue:
Proceedings of the fourth workshop on High performance computational finance
Year:
2011

Citing 9
Cited 0

Efficient and portable combined random number generators

Communications of the ACM
Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator

ACM Transactions on Modeling and Computer Simulation (TOMACS) - Special issue on uniform random number generation
Multi-paradigm design for C++

Multi-paradigm design for C++
Control flow analysis

Proceedings of a symposium on Compiler optimization
Introduction to C++ for Financial Engineers with CD-ROM: An Object-Oriented Approach (The Wiley Finance Series)

Introduction to C++ for Financial Engineers with CD-ROM: An Object-Oriented Approach (The Wiley Finance Series)
C++ Design Patterns and Derivatives Pricing

C++ Design Patterns and Derivatives Pricing
Acceleration of market value-at-risk estimation

Proceedings of the 2nd Workshop on High Performance Computational Finance
Cost vs. performance of VaR on accelerator platforms

Proceedings of the 2nd Workshop on High Performance Computational Finance
Monte Carlo Frameworks: Building Customisable High-performance C++ Applications

Monte Carlo Frameworks: Building Customisable High-performance C++ Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we are examining the impact of modern programming language abstractions on total cost of ownership (TCO) of a financial computing operation. Our analysis is based on static and dynamic analysis of example financial software, based on our loop-flow graph (LFG) concept and our custom dynamic hotspot tool called MaxSpot. Our results show that, if the required throughput of an application is high enough, then operational expenditure is minimized by minimizing runtime and not programming effort.