We use cookies on this website to ensure that you get the best user experience. By continuing to use this website you are consenting to the use of these cookies. To find out more please follow this link

Pure FPGA Market Data Solutions

Fan-out and Data Distribution

Fan-out and Data Distribution

Fan-out and Data Distribution

The NovaSparks pure FPGA processing of market data includes the data distribution, performed in FPGA as well.

Messages (including only the relevant information for the users) are duplicated and filtered before being delivered to a high number of PCIe or Ethernet interfaces.

The performance of the feed handler remains unaffected whether each message is sent out to one interface or twenty. Thanks the parallelism capability of FPGA and the NovaSparks FPGA Matrix architecture, a message can be replicated as many times as there are output interfaces in just one clock cycle.

The same activity in a CPU based architecture requires to copy the message as many times and creates high latency spikes, especially during market data peaks.

Coupled with instrument and exchange filtering, this capability allows NovaSparks customers to customize a dedicated output stream for each of their trading applications, without impacting the overall latency of the ticker plant.

Key Advantages

  • The NovaSparks feed handlers include a fast market data distribution capability
  • Up to 12 PCIe interfaces and tens of Ethernet interfaces are available per appliance
  • Multiple trading servers can consume specific market data streams in parallel without impacting the overall processing latency of the ticker plant

© 2015 Copyright. All rights reserved.

Website proudly manufactured in the UK by DigitalParentCompany.com

NovaSparks logo