Commentary

Jared Dillian
Traders Magazine Online News

Was it Worth It?

In this piece from 10th Man, author Jared Dillian discusses how the ETF revolution is less about ETFs and more about indexing; about how people have come to view stocks less as stocks and more as blobs of stocks.

Traders Poll

Would you feel better if the Chicago Stock Exchange were purchased by U.S. firm or consortium rather than a foreign one?

Yes

73%

No

4%

Doesn't matter to me

23%

Free Site Registration

August 14, 2007

Gentlemen, Start Your Engines

Processing More Data with Less Latency

By Nina Mehta

Also in this article

The official theme of this year's annual Securities Industry and Financial Markets Association technology show-"Tech Be Nimble! Tech Be Quick!"-was borne out by a tour of the exhibition halls. With speed becoming increasingly critical to trading-related applications, dozens of vendors emphasized their low-latency, high-throughput services for buyside and sellside firms.

Altogether, some 300 vendors showcased their wares at SIFMA's Technology Management Conference and Exhibit. About 7,000 financial industry attendees made the rounds over three days.

The need for speed and low-latency applications has been driven, in large part, by the Securities and Exchange Commission's Regulation NMS, which fueled a surge in automated and algorithmic trading. With more automated trading, firms must be able to process and distribute the larger volumes of incoming market data to various internal applications simultaneously and make trading decisions faster.

The goal is to shrink latency, or time delay, wherever it exists, whether it's in market data transmission, network communications or execution-related processes. And the contest, in some cases, is now registered in microseconds, not just milliseconds. A microsecond is one-thousandth of a millisecond, or one-millionth of a second.

The granddaddies of complex event processing (CEP) were all at the tech show. Progress Software's Progress division, StreamBase Systems, Aleri and others showcased their offerings for an increasingly speed-sensitive audience. Event processing refers to the high-speed search for patterns or "events" in vast streams of data that can be used to automatically generate trading decisions, or for other activities such as trade monitoring and compliance.

CEP News

Perhaps the most surprising application for commercial CEP software comes from the U.K.'s Financial Services Authority. The FSA will use Progress Software's Apama event processing platform to provide real-time fraud detection capabilities starting early next year.

The FSA's new Surveillance and Automated Business Reporting Engine (Sabre II) is the first project, said John Bates, founder and vice president of Progress Apama, in which a regulator will look for illegal or potentially harmful complex trading patterns in real time. This capability will allow the FSA, if warranted, to potentially take action before those events move the market.

The FSA already has a portfolio of hundreds of patterns it intends to look for in trading data. FSA staff will also be able to readily "open up new patterns they want to look for or create new scenarios they want to monitor," Bates said. They can compare trading events as they occur to similar past events, including those involving the same traders, and can customize alerts for specific traders or firms. The regulator will launch this new monitoring effort with equities.

StreamBase Systems released StreamBase 5.0 at the SIFMA tech conference. The CEP platform enables customers to build applications faster and to query, process and analyze real-time and historical data at a rate of hundreds of thousands of messages per second. The platform supports more advanced order types, various trading and execution strategies, and real-time transaction cost analysis.

Integrated Solutions

Software company Sybase also announced that its Real-Time Analytics Platform now integrates with StreamBase's CEP platform. Firms with large data storage requirements can use the joint solution for back-testing algo strategies and for risk analysis.