About a-team Marketing Services
The knowledge platform for the financial technology industry
The knowledge platform for the financial technology industry

A-Team Insight Blogs

Exegy Introduces FPGA-Based FIX Protocol Support to Enhance Performance and Efficiency

Subscribe to our newsletter

Front-office trading solutions provider Exegy has integrated FIX protocol support into nxFramework, its FPGA-based development framework, aiming to reduce server resource consumption for FIX protocol processing by an estimated 10-20 times.

In May last year, Exegy acquired Paris-based Enyx, a leading developer of FPGA-based, high-performance trading solutions. By leveraging their extensive expertise in FPGA technology, Exegy has now successfully offloaded FIX protocol processing to the FPGA in a strategic move to not only ensure lower latency but to also substantially decrease server load, thereby freeing up CPU cores for more critical business functions.

Now over thirty years old, the FIX protocol has become a cornerstone of electronic messaging within financial markets. Despite its flexibility and extensibility, the protocol’s design can be resource-intensive, particularly when processing high volumes of FIX traffic through conventional software-based architectures.

“FIX is a flexible and widely used protocol that is very compute intensive for standard CPU-based architecture,” Laurent de Barry, a co-founder of Enyx and now Senior Director, Global Head of Solutions Consulting at Exegy, tells TradingTech Insight. “As data volumes continue to grow and data centre space becomes scarcer, we wanted to offer our customers an alternative to process FIX traffic more efficiently. This new FIX engine offloads the protocol processing to the FPGA, therefore freeing up valuable CPU cores for the business logic and reducing total cost of ownership.”

Exegy’s FIX engine utilises FPGAs to manage hundreds of sessions using just two CPU cores, a task that would typically occupy a least ten times as many cores using a pure software-based approach, according to the company. By employing pre-configured templates to validate FIX messages, filtering out superfluous data and offloading substantial processing workload from software to FPGA, the design enables firms to allocate their computational resources to more productive user applications rather than FIX protocol processing.

“We acknowledge that our customers operate a large software infrastructure that is vital to their business. With this solution, our goal was to offer an easy, painless way to enhance these software applications allowing the hardware to take on the resource intensive task of processing the FIX protocol,” says de Barry. “Customers can now free up valuable CPU cores for their business logic by running the FIX engine on a relatively inexpensive FPGA card without needing to upgrade or add new servers.”

By bridging the divide between hardware and software, Exegy’s solution is designed to provide a software-configurable, FPGA-based FIX stack that offers deterministic manipulation of data streams with low latency, on-the-fly data content modification, and retransmission of altered FIX messages within the FPGA across multiple sessions. Additionally, it promises reduced power consumption and lower operational costs.

The framework also includes a comprehensive suite of software libraries for configuration, monitoring and support, facilitating seamless integration with existing systems. Dedicated APIs are provided for controlling the TCP stack, configuring the FIX offload engine, and managing FIX sessions.

“We focused on providing an FPGA-powered solution, tailored for software users,” says de Barry. “To ensure seamless integration, nxFramework handles each aspect of the FPGA connectivity (i.e. TCP, UDP, PCIe) with simple configuration APIs for the software application, allowing the user to focus on their custom processing and core logic.”

Subscribe to our newsletter

Related content

WEBINAR

Recorded Webinar: How to move to a modern, component based trading architecture using a Buy AND Build approach

To remain competitive in today’s electronic markets, firms need trading architectures that support rapid innovation, effortless integration of new capabilities, and the agility to respond to shifting market demands. This is prompting technology leaders to move beyond the traditional “Buy vs. Build” debate, a false dichotomy that oversimplifies the choice between generic, off-the-shelf platforms and...

BLOG

BitGo and Susquehanna Build First Institutional OTC On-Ramp to Prediction Markets

Prediction markets have attracted growing institutional interest as tools for price discovery around political, economic, and event-driven outcomes, but participation has remained largely confined to retail platforms with workflows that most institutional trading desks cannot operationally support. BitGo and Susquehanna Crypto are now attempting to close that gap with what they describe as the first...

EVENT

TradingTech Summit London

Now in its 15th year the TradingTech Summit London brings together the European trading technology capital markets industry and examines the latest changes and innovations in trading technology and explores how technology is being deployed to create an edge in sell side and buy side capital markets financial institutions.

GUIDE

Regulatory Data Handbook – Second Edition

Need to know all the essentials about the regulations impacting data management? A-Team’s Regulatory Data Handbook is a great way to see at-a-glance: All the regulations that are impacting data management today A description of each regulation The impact each will have from a data and data management perspective Messages from sponsors with products related to...