Truviso Algorithmic Platform Uses SQL-Based Event Processing
November 1, 2007
Foster City, Calif.-based Truviso is emerging from stealth mode with the release of a platform for the development, testing, execution and evaluation of proprietary algorithmic trading strategies. The product’s complex event processing (CEP) engine is based on a SQL database standard that the company says is superior to competing CEP technologies.
The platform, Algorithmic Trading Solution, which combines an event processing engine with real-time and historical data analysis, is targeted to hedge funds, who Truviso executives believe are more likely to seek tools built on familiar development languages such as SQL to simplify the process of testing and adjusting a large number of algo trading strategies.
“Our goal is to offer an algorithm trading platform that is complete, can change and cope with the accelerated pace of the markets, and yet leverages the skills that exist within the IT organization of a hedge fund,” said Boyd Pearce, chief executive of two-year-old Truviso, which was known until recently as Amalgamated Insight.
The platform’s event processing engine continuously executes hundreds of SQL queries against multiple real-time market-data feeds, according to the company. The Algorithmic Trading Solution also facilitates historical trend analysis and real-time algorithm execution in a single framework, which it claims provides a wider range of analysis capabilities than are available in other CEP solutions.
“Many hedge funds don’t have the expertise or the people trained on each of the platform components to be able to build a fully integrated, fully combined offering such as ours,” said Pearce.
“Hedge fund traders are constantly looking to improve their algorithmic strategies,” he added. “Most algorithms are only truly differentiated for a few weeks or months at a time,” but “our environment gives you the ability to constantly install changes, do data-mining, back-testing, running in parallel--and do all of this at an accelerated pace.”
Not all providers of event processing technology are fond of SQL. A blog from Progress Software Corp.’s Apama division says that although a number of new vendors are advocating the use of SQL, they believe it is not the ideal way to build CEP-based applications. “SQL is certainly well understood,” says the blog, “but by providing an SQL interface to event processing products, practitioners assume that an SQL way of thinking will be appropriate. It isn’t. By thinking of event processing as actually a real-time database you get stuck in a database-centric design pattern.”
Truviso co-founder and CTO Michael Franklin, a professor of computer science at University of California at Berkeley and a well-known researcher in the database field, is a supporter of PostgreSQL--an open-source version of SQL that has gained a following among developers as well as less-technically proficient users of technology--which the company uses for its event processing.
According to Franklin, CEP vendors such as Aleri, Coral8 and StreamBase Systems are using SQL or SQL-like languages to build applications, but he believes PostgreSQL is richer and more advanced. “PostgreSQL has been around since 1995 and while other vendors utilize subsets of SQL or languages that look like SQL, none can run full SQL queries as you can on PostgreSQL,” he said.
Pearce said that Truviso will charge customers a subscription fee based on the volume of business they conduct. “A smaller fund would pay less, depending on their trading volume,” he said. “Fees will range from as low as a few thousand a month to tens of thousands if you also choose to use additional features such as our visualization tools.”
