SR Labs Gears Up for Collocation

Several weeks after announcing connectivity to the BATS Trading ECN platform in early July, electronic trading system developer SR Labs is setting its sights on competing in the market data collocation space.

"We don't offer collocation services ourselves. We provide software that is optimized for collocation," says Samir Islam, VP of software development at SR Labs, developer of the Low-latency Order Management System known as LOMS. "Collocation is something we're exploring. It's not something we're doing yet," says Islam.

SR Labs — whose full name is Street Response Laboratories — has written an SMDS (SR Market Data Solution) line handler to the BATS' Pitch feed. "We're able to pull in market data at a very high speed from BATS so that clients can execute off of it," says Islam, a former VP at Merrill Lynch who had experience building applications for risk management, pricing and high-frequency trading in the equity derivatives business.

SR Labs has built market data handlers for between 15-to-20 exchanges, and supports Level II price points and trade information, according to Islam. Currently, all North American exchanges are supported, including NYSE ArcaBook (both equities and options), Nasdaq Total View, CME and OPRA FAST (FIX Adopted for Streaming). Internationally, the firm is also supporting Chi-X, Deutsche Borse's Xetra, Eurex FAST, London Stock Exchange, and NYSE Euronext and Tokyo Stock Exchange. It has plans to add Turquoise, BATS Europe, Sydney, Bovespa and Dubai.

According to Islam, SMDS is geared for traders — usually proprietary groups or quantitative traders — that execute robotic trading applications, which essentially trade by themselves. The time between the trade happening and the market data application receiving that information needs to be as minimal as possible, he says. "One thousandth of a second will cost slippage and cause them to lose money. We're talking microsecond latency for these applications to work," says Islam.

Most of the exchanges provide a market data feed for collocation, which is in their native format. What SMDS does is take the data from all the exchanges it has written to, normalizes it to a single format and hands it over to the application. "The advantage is that since an exchange has a different format for its data, if you are trading on NYSE and now want to use the same application on Nasdaq, you'd have to rewrite your applications," says Islam. "Even though the application is the same, the way it gets market data from each exchange is very different," he says. Potentially, clients could receive market data from any of those exchanges without changing their trading applications, he says. "We do it without adding additional latency to their trading application. There has to be some, but we're very competitive," says Islam, who notes the firm's competitors in the low-latency market data aggregation space would be ACTIV Financial Software, Exegy and Skyler Technology.

The SMDS software would be collocated in an exchange data center, and run on the same hardware as the client's trading application runs on. "With most of our competitors, the client 's trading application would run on another box, which would be next to the line handler taking in data from the exchange to parse it and normalize it," he says. "We run within the same process as the client applications and I believe most of our competition does not," says Islam. Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio