When deterministic latency is critical, the goal is to avoid the bottlenecks inherent to CPU-based implementations and to maximize the use of FPGA’s even though the data is coming from multiple markets. Designed for time-sensitive trading strategies such as Market Making, Statistical Arbitrage or Algorithmic Trading, the NovaSparks platform adopts a pure FPGA architecture engineered to unleash the highest speed and enable trading firms to maximize the returns on their strategies.
This pure FPGA approach uses an expandable matrix of FPGA nodes that are linked with raw binary interconnections.
In addition, to avoid bottlenecks that can occur with other system resources, each FPGA node comes with its own set of Inputs/Outputs and memories. This modular matrix allows the system to grow proportionally across multiple market feeds.
As a result, all functions of the market data parsing, book building, filtering and distribution can be done in hardware without bottlenecks, irrespective of the number of feeds received, of the rate of data or the number of downstream consumers.