Palo Alto (18 Sep 2006) – Kx Systems, leader in high-performance databases and timeseries analysis, today announced that Cantor Fitzgerald has completed a comprehensive system to handle historical data analysis and immediate querying of streaming market data based entirely on the Kx kdb+ database and “q” programming language. Cantor’s traders use the solution to integrate their risk management and trading strategies and optimize short-term opportunities.
With this system, traders now track their portfolio risk and trades in realtime, in relation to other trades being made in the market, thus immediately gauging the momentum of price moves. This capability enables Cantor to make trading decisions instantly and get in and out of a market quickly — without compromising risk management.
“Everyone talks about realtime risk management, but with kdb+ we actually attain it,” said Jacob Loveless, SVP at Cantor. “We’ve gone beyond simply knowing the risk to being able to act on it in the moment.”
Cantor began working with Kx software in November 2005 and wrote a back-testing program in two days. Other capabilities were then added to build on the uniform kdb+ architecture for historical and real-time data. “The ability to integrate realtime publishing with historical searches made kdb+ the obvious choice for our new analysis system,” said Loveless.
Because the entire system is written in the q language integrated with kdb+, there is zero latency between data capture and analysis. The unified platform architecture used by Kx eliminates the latency found where traditional applications connect to multiple external databases, and their various query languages, as well as to feed handlers and servers written in a variety of languages.
“Kx provides a unified platform that is unmatched for flexibility,” added Loveless. “In this industry, our priorities for a programming language are simple – productivity and speed. Q delivers both, and it includes a realtime database that can keep up.” kdb+ eliminates the typical development scenario in which programmers must update several often incompatible databases and programs in order to develop, test, and release new code. For Cantor making changes to different parts of the system is fast and simple.
Cantor has also taken advantage of kdb+ support from Kx Systems’ sales and services partner First Derivatives (FD), plc, “Support response using FD’s email support system is fast,” said Loveless. “Typically, I email a question, go get a cup of coffee, and have the response I need waiting for me when I get back to my desk.”
“Cantor is the ideal example of how the world’s most innovative investment firms are using Kx Systems products today,” commented Simon Garland, Kx CTO. “It’s no longer about building separate systems for tick data, historical queries, cross-asset data trading, and so on. It’s about solving the entire puzzle at once, which is exactly what kdb+ was designed to do.