Wall Street & Technology is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Trading Technology

12:04 PM
Connect Directly
RSS
E-Mail
50%
50%

IBM Upgrades Direct, Low-Latency Delivery of Exchange Data

In their ongoing quest to reduce data latency in the delivery of market data to algorithmic trading programs, Wall Street firms are starting to take data feeds directly from the exchanges, cutting out middlemen like Reuters, Thomson and Bloomberg. IBM's recently-released WebSphere Front Office for Financial Markets 2.0 is one of the few products (Wombat offers another) that can accommodate this. "Customers have told us they're interested in having a platform that can handle data that isn't neces

In their ongoing quest to reduce data latency in the delivery of market data to algorithmic trading programs, Wall Street firms are starting to take data feeds directly from the exchanges, cutting out middlemen like Reuters, Thomson and Bloomberg. IBM's recently-released WebSphere Front Office for Financial Markets 2.0 is one of the few products (Wombat offers another) that can accommodate this. "Customers have told us they're interested in having a platform that can handle data that isn't necessarily tied to a data provider, says Larry Baldauf, director of financial services strategy for IBM's software group. "We can connect to different exchanges and data sources and distribute the data to applications."The product has about a dozen prebuilt connectors to major North American and European exchanges, and Baldauf says IBM has an aggressive plan to lengthen that list. The software accepts the exchange data and puts it into a common format that can be read by trading applications.

What kind of latency improvements can firms expect to see? "We've measured latency on our platform of under 200 microseconds," Baldauf says. "Typically in the industry, millisecond latency has been the target, so we're well under that. In head-to-head tests versus other competitors in this space, we've had superior latency and throughput."

The software can run on Ethernet or Infiniband networks - Infiniband support is new with this release. "Infiniband does speed things up considerably," Baldauf says. Also new is support for.net and Solaris clients as well as newer versions of Linux, such as 64-bit Linux. IBM has added Unicast and remote office support, which allows firms to deliver data beyond a local area network to remote offices and clients, a feature Baldauf says clients have asked for.

IBM also announced recently that it's offering the underlying low-latency messaging technology in WebSphere Front Office for Financial Markets as a separate product, called IBM WebSphere MQ Low Latency Messaging Version 2.0. "This came out of IBM's Haifa research lab, which came up with unique ways of packetizing and transmitting the data for superior performance," Baldauf says. This messaging technology could be used to speed data delivery, for instance, to reference data, risk and compliance, event processing and analytics software. According to IBM, it can can deliver approximately one million 120-byte messages per second on Ethernet, close to three million 120-byte messages per second on InfiniBand, and more than eight million smaller messages per second, all on common x86 servers. In tests it has delivered latency of 30 microseconds for 120 byte messages delivered at 10,000 messages per second on InfiniBand or 61 microseconds on Ethernet. Reuters will be OEMing this IBM data transport in its forthcoming multicast server next year.

Register for Wall Street & Technology Newsletters
Video
Exclusive: Inside the GETCO Execution Services Trading Floor
Exclusive: Inside the GETCO Execution Services Trading Floor
Advanced Trading takes you on an exclusive tour of the New York trading floor of GETCO Execution Services, the solutions arm of GETCO.