Hardware Acceleration & HPC
Ultra-low latency Ip cores for hardware acceleration of a trading platform
The Challenge
Develop ultra-low latency and ultra-low gate count Ethernet MAC/PCS and RS-FEC IP Cores for algorithmic trading.
CLIENT: Investment Banks & HFT Firms
.
LOCATION: Worldwide
As investors have always been looking for new and innovative ways to gain an advantage over other market players, our client asked Orthogone to develop ultra-low latency, ultra-low gate count Ethernet MAC/PCS and RS-FEC IP cores to handle high-frequency transaction requests between the bank and the stock exchange, thus differentiating itself from the competition.
Market Data Feed Handler and Tick-to-Trade application
Network Security Appliances
High-Performance Computing
Experience Unparalleled Speed and Efficiency
PRODUCT BRIEF - Ultra-Low Latency 10G MAC & PCS.
Download our product brief and unlock the power of high-speed, efficient Ethernet solutions for optimized system performance.
The Solution
Establish a partnership with Orthogone to develop a customized solution
Orthogone used a holistic approach to develop multiple Ethernet MAC/PCS flavours to support multiple data rates and Forward Error Correction (FEC) options. A single Verilog source code platform was used to support all configurations, options, and technology process nodes. This greatly simplified test, integration and regression and preserved a uniform design for all options and configurations.
Ultra-low latency Ip cores for hardware acceleration of a trading platform
Multiple design innovations in the data processing algorithms resulted in ultra-low gate count and impressive latency performances making the cores ideal for applications where latency is critical (e.g., algorithmic trading).
TECHNOLOGY
- FPGA: Xilinx (Virtex-7, UltraScale and UltraScale+)
- FPGA: Intel (Stratix-V, Arria-10, Stratix-10)
The Result
Implemented a platform to manage high frequency trading
The solution targets ultra-low latency execution, high transaction throughput and robustness through transmission redundancy and data compression to maximize bandwidth. Multiple deployments have been completed in different market segments and geographic locations.
