Articles Marketmedia

Using Hardware to Speed Up Pre-Trade Risk Checks

Written by Terry Flanagan | Sep 13, 2012 7:59:06 AM

Trading platforms are embedding specialized hardware capable of addressing industry pain points such as pre-trade risk, compliance and ultra-low latency execution.

The hardware, in the form of appliances called FPGA (field-programmable-gate-array), are designed to enforce pre-market risks in nanoseconds, while preventing impact on latencies that historically have slowed down traditional trading systems and models.

“The industry has always had to make choices between speed and risk,” said Matt Dangerfield, chief technology officer at Fixnetix, the U.K.-based co-location service provider. “Choices are a thing of the past.”

Specialized microchips employing FPGA have near zero impact on latency for governing required checks directed by the Securities and Exchange Commission’s Rule 15c3-5 and similar regulations from other regulatory governing bodies.

Fixnetix and NovaSparks have launched a managed FPGA market data service for exchange co-location venues in the U.S. and Europe.

The service uses the NovaSparks FPGA Market Data Matrix feed handlers, combined with Fixnetix iX-eCute FPGA Market Gateways for execution services.

“The entire market place has been seeking a common technology platform for both market data and trading coupled with the 24/7 managed service component,” said Dangerfield. “FPGA market data coupled with FPGA trading systems [iX-eCute] supported by managed services allows both companies to bring a ‘one stop’ shop.”

Two major North American equities exchanges, NYSE Euronext and Nasdaq, have expanded their risk management offerings through acquiring or investing in established technology providers, Fixnetix in the case of NYSE Euronext and FTEN in the case of Nasdaq.

“The players in the lowest latency space, such as FPGA and microwave, are highly fragmented, with production offerings coming from smaller, niche providers—while larger network and telco providers are slower to figure out their strategic moves in the low latency technology space,” said Emmanuel Doe, president of trading solutions at Interactive Data Corp, a provider of outsourced ultra-low latency trading infrastructure.

Fixnetix specializes in managed services for market data, trading and risk control, while NovaSparks provides FPGA-based feed handlers technologies to with nanosecond and deterministic (consistent) market data processing speeds.

“The relationship is based on a mutual company ethos/operating model which is underpinned by common technology to provide a managed service utilizing Nova Sparks technology,” said Dangerfield at Fixnetix. “Nova Sparks’s market data technology is second to none within the vertical, providing ultra-low latency feeds for Tier One market protagonists.”

Nova Sparks’ Gen2 feed-handler, pushes data from multiple sources to multiple servers using uniquely engineered FPGA-technology.

When processing the Nasdaq TotalView feed, average latency performance was 400 nanoseconds with no data points being above 900 nanoseconds, which are the fastest speeds and the most deterministic in terms of market data processing, the company said.

At the servers at the endpoints, the operating system itself contributes tens or more microseconds to the latency budget of a financial transaction.

Operating system (OS) bypass and direct memory access techniques can reduce latency to single digit microseconds by giving the application direct access to the network.

OS bypass is a form of hardware acceleration by which an application can directly access input/output commands without invoking the operating system, enabling, the CPU to avoid having to switch between the application and OS, a process known as “context switching”.

It’s estimated that up to 40% of network overhead is attributable to context switching.
Solarflare’s OS bypass solution, OpenOnload, is an “accelerated network stack” that provides applications with direct access to network hardware.

The network stack bypasses the operating system and significantly improves performance through the removal of context switches, which otherwise reduce the efficiency by which a processor can execute application code, according to Solarflare, a software and hardware provider.

This acceleration is most desirable for applications that are network-intensive, such as market data and high-frequency trading applications.