The goal of this project is to rapidly construct an order-by-order level snapshot of financial markets with nanosecond resolution time stamps. We are particularly interested in understanding the impact of high-frequency traders on the security, stability, and fairness of these markets. In this paper, we describe our computational approach, the optimizations that were applied to improve the performance of the software by a factor of more than 125×, and the new capabilities that can be enabled using a combination of fast algorithms and XSEDE resources.
- high-frequency trading
- performance optimization
ASJC Scopus subject areas
- Theoretical Computer Science
- Computer Science Applications
- Computer Networks and Communications
- Computational Theory and Mathematics