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

Zeptonics Taps Mellanox For Ultra-Low-Latency Matching Engine

Subscribe to our newsletter

Sydney-based Zeptonics is integrating network interface cards from Mellanox Technologies into its ZeptoMatch matching engine. The integration boasts wire-to-wire matching for a million orders per second with a median wire-to-wire match time of 1.97 microseconds and a typical client roundtrip of three microseconds.

ZeptoMatch is a software-only application, running on commodity x86/Red Hat Linux servers, with solid state storage. Mellanox ConnectX-3 EN 10gE network cards support network to application data flow using kernel bypass. “We developed all the software components ourselves, but even though we have considerable hardware design expertise in-house, we decided to use the Mellanox cards for the physical interface, and that has definitely helped us deliver a new low-latency benchmark for trade matching,” says David Snowdon, principal at Zeptonics.

Both FIX and an Ouch-based binary protocol are supported for order execution and an embedded firewall ensures the integrity of data packets. The matching engine can also be paired with ZeptoRisk, for pre-trade risk checking as increasingly required by regulators.

ZeptoMatch might typically be integrated with networking devices from Zeptonics.  ZeptoMux is a 130-nanosecond layer 2 switch for in-bound multiplexing; and ZeptoLink, is 5-nanosecond circuit switch for fan-out of market data.

Subscribe to our newsletter

Related content

WEBINAR

Recorded Webinar: From Data to Alpha: AI Strategies for Taming Unstructured Data

Unstructured data and text now accounts for the majority of information flowing through financial markets organisations, spanning research content, corporate disclosures, communications, alternative data, and internal documents. While AI has created new opportunities to extract signals, many firms are discovering that value is constrained not by models, but by the quality of the content, architecture,...

BLOG

DiffusionData Targets Agentic AI in Finance with New MCP Server

Data technology firm DiffusionData has released an open-source server designed to connect Large Language Models (LLMs) with real-time data streams, aiming to facilitate the development of Agentic AI in financial services. The new Diffusion MCP Server uses the Model Context Protocol (MCP), an open standard for AI models to interact with external tools and data...

EVENT

AI in Data Management Summit New York City

Following the success of the 15th Data Management Summit NYC, A-Team Group are excited to announce our new event: AI in Data Management Summit NYC!

GUIDE

Valuations – Toward On-Demand Evaluated Pricing

Risk and regulatory imperatives are demanding access to the latest portfolio information, placing new pressures on the pricing and valuation function. And the front office increasingly wants up-to-date valuations of hard-to-price securities. These developments are driving a push toward on-demand evaluated pricing capabilities, with pricing teams seeking to provide access to valuations at higher frequency...