11.21.2014
By Terry Flanagan

Stakes Higher for Surveillance

The stakes are getting higher for lapses in operational risk, with banks getting hit with multimillion fines and penalties for systemic failure to catch rogue traders.

“We are getting perhaps less tolerant of the way these traders behave,” Giles Nelson, vice president of product strategy, intelligent business operations at Software AG, told Markets Media. “What has happened, therefore, is that this has caused significant pressure upon firms to really change the way they do market surveillance.”

Giles noted that there are not only ‘fat-finger’ errors, but also those that involve premeditation. “We had the London Whale, SocGen and most recently the wide-spread evidence of foreign-exchange fixing at Royal Bank of Scotland, JP Morgan, Citi, et al.,” he said.

Five years ago, it was still acceptable for a bank to be able to report on the way its traders were operating, and perhaps those of its clients’ trading as well, on a next-day or end-of-day. “Now that is no long sufficient,” said Nelson. “You need to have real-time reporting, real-time monitoring, and that will not only be able to catch people more quickly, and that of course is important because you don’t want the market moved by this kind of behavior, but it will also act as much more of a deterrent.”

Software AG’s latest release of Apama Streaming Analytics includes the addition of new high-bandwidth channels, which the company says boosts performance twenty-fold on a single computer producing upwards of 32 million events per second; in a scaled-out architecture, performance can increase to billions of events per second.

Software AG’s customers are using Apama for market surveillance. “They’re looking at orders placed, they’re looking at text from chat rooms, for example,” said Nelson. “And trying to determine then when traders are behaving inappropriately and reporting on that.”

The higher performance is achieved by splitting a data stream into multiple, separate channels to exploit the power of parallel processing pipelines whenever possible.

“What Apama is at the heart is a software engine for making real-time event-based decisions,” Nelson said, “When you’re dealing with markets where you’re wanting to make decisions upon the information very quickly, then this kind of software is the right thing to use because you can look up more information more quickly than you can using conventional event processing databases.”

The higher processing capability means developers can structure their applications in such a way that the data can be processed completely separately and executed in parallel, according to Software AG.

“It gives a lot more scalability, a lot more performance, so on commodity hardware our customers can deploy Apama to be able to consume and analyze more data,” Nelson said. “When you’re talking about an industry that does, of course, have a lot of data, but also where you’re wanting to introduce new channels to be able to analyze the data from unstructured social media, chatroom data, and not just conventional order and trade data, then having that extra performance is very important.”

Featured image by viappy/Dollar Photo Club

Related articles

  1. Agency broker moves beyond execution to offer a broader suite of services.

  2. Algorithms have become more prevalent in the spot FX market.

  3. Congress Unlikely to Act on HFT

    QB’s Algo Suite for futures market trade execution is also being co-located to HKEX.

  4. Breaking data silos is key to deploying automation beyond 'nuisance' orders.

  5. They can be used on quantum hardware expected to be available in 5 to 10 years.