Hardware Appliances Power Ticker Plants09.21.2011
FPGA is deployed to boost market data feeds and drive down latency.
NovaSparks has announced that Gen2, an ultra-low latency market data feed-handler, performed with average latency at 400 nanoseconds during the high volatility market conditions of August.
The NovaSparks Gen2 stand-alone appliance is a FPGA (field-programmable-gate-array) solution and is not a hardware accelerated solution. All feed handling is performed within dynamically customized hardware without the need for a traditional generalist CPU.
“The NovaSparks Gen2 feed-handler uses 100 percent FPGA technology,” Yves Charles, CEO of NovaSparks, told Markets Media.
Other CPU or hardware accelerated solutions cannot yield such latencies because bottlenecks are created when software meets hardware, Charles said. NovaSparks guarantees sub-microsecond deterministic latency because all latency reduction is achieved using FPGA programming which cuts through bottlenecks.
“In times of market spikes what sets trading firms apart is their ability to execute orders faster, better and at the best price,” Charles said. “The characteristics of FPGA programming ensure that information is guaranteed to be delivered at the required speed unlike other hybrid solutions that cannot be deemed guaranteed or deterministic.”
The Gen2 feed-handler, offers scalable, adaptable and fast FPGA feed-handler, ticker plant and order book management system. Its fan-out capabilities push data from multiple sources to multiple servers using uniquely engineered FPGA-technology, said Charles.
When processing the NASDAQ TotalView feed, average latency performance was 400 nanoseconds with no data points being above 900 nanoseconds; these are the fastest speeds and the most deterministic, in the market today in terms of market data processing.
“NovaSparks tested the Gen2 feed-handler during August, a period of great market volatility,” Charles said. “The outstanding latency figures reported have proved its deterministic capabilities; producing the same latencies whatever the market conditions, whatever the number of markets being monitored and whatever the market volumes.”
Exchange group reported its best second quarter, following a record first quarter.
Institutions can manage their bitcoin exposures in existing portfolio management and trading workflows.
Enabling a straight through process reduces operational complexity.
The new zero-footprint datafeed removes the need for dedicated market data infrastructure and equipment.
A robust sector taxonomy helps recognize similarities between digital assets which lead to correlated returns.