Stream-processing technology is big news in the market data world, where the demands of algorithmic trading are driving investment in solutions to support high-volume, low-latency, fast-moving environments. To date, it has not been widely suggested that low latency was so important in the context of reference data, but an early sign that streaming technology could be about to make the leap from market to reference data comes from one large, as yet unnamed, bank that is deploying the Aleri Streaming Platform for reference data distribution.
Aleri president and CEO Don DeLoach says that though Aleri is involved in a couple of different projects “with workflow aspects”, “a securities master file implementation is an intriguing use” of its streaming platform. More typical applications of the technology are in the areas of algo trading, MiFID (the vendor has developed a best execution prototype, MiFID Monitor, October 2006), position management and pricing.
“If you think of stream processing, you usually think of it for trading and algo trading. You wouldn’t usually think of it for something as hardened and extensive as reference data,” he says. “But someone already familiar with the product said they wanted to use it for reference data.”
Aleri is working with the bank to distribute reference data, using the streaming platform to manage the state of numerous different silo systems, and distribute the data according to a number of rules based on the state of those underlying silo solutions, DeLoach says. “This makes the process much more automated. The application is unique, because there is capacity in there to know the state of the different silos; it’s more than just moving messages around. It’s very much a service app-roach, with each underlying system communicating with the Aleri layer.”
According to Nick Deacon, senior vice president at AleriLabs – the banner under which the streaming platform is provided – the key driver for the bank’s deployment of the Aleri technology for reference data is speed. “There is so much data to deal with, and the ability to synchronise it across systems is important,” he says. The data can also be complex, he adds. “The message may go through Aleri, and then be impacted by something happening in another system. Because the Aleri layer knows that, it can determine what to do with the message as a result.”
DeLoach says Aleri is in discussions with other banks about deploying the streaming platform for reference data. “We won’t productise the solution for reference data specifi-cally, but I do anticipate us adding capabilities to enhance its use in a reference data environment,” he says.