In the fast-paced business of electronic securities trading, each nanosecond counts. A new, High-Frequency Trading (HFT) solution, jointly developed by Penguin Computing®, Solarflare®, LDA Technologies®, and XILINX®, has been nicknamed the Ultimate Trading Machine, as it sets the bar for tick-to-trade performance. On 17 October 2017, at the STAC Summit in Chicago, IL, it was announced that this Ultimate Trading Machine achieved 98 nanosecond tick-to-trade latency, making this solution 18% faster than the previous world record of 120 nanoseconds achieved in May 2017.
The time interval between receiving a market tick and sending a buy or sell order is the tick-to-trade latency. Decreasing the tick-to-trade latency improves the queue positions of trades. This results in higher probabilities for trades to be executed, and, thus, delivers a competitive advantage.
The solution was evaluated by the Securities Technology Analysis Center (STAC®). STAC coordinates the STAC Benchmark Council, a community that consists of over 300 financial institutions and more than 50 vendor organizations. The purpose of the Council is to explore technical challenges and solutions in financial services and to develop technology benchmark standards that are useful to financial organizations, such as banks, brokerage houses, exchanges, proprietary trading shops, and other market participants. STAC evaluations are respected throughout the industry.
The Ultimate Trading Machine was evaluated using the STAC-T0TM benchmark suite. This benchmark simulates the time for the Chicago Mercantile Exchange to emit a trade data packet and for the Exchange to receive a market order packet. This is the minimum, possible trade execution time. STAC-T0 is designed to isolate the tick-to-trade network I/O latency and to measure this latency with high precision (an uncertainty of just ±2 nanoseconds is incorporated into reported results).
The solution achieves ultra low processing latencies through optimal design choices and select components. It is built on a solid performer, the Penguin Computing Relion® XE1112 server. This 1U, 19″ EIA form-factor server features dual Intel® Xeon® Scalable processors and optimized memory, bus, and storage options (including low-latency, Intel OptaneTM data center storage). The interconnect is powered by the Solarflare XtremeScale 8722 software-defined network interface, delivering ultra-low latency and onload kernel bypass with offload to a Xilinx Kintex® UltraScaleTM FPGA running the LDA Technologies’ Lightspeed TCPTM core software stack.
The 98 nanosecond result was achieved at 1 gigabit per second ingress for 68-byte frames and 109 nanoseconds for 507-byte frames. This is the maximum actionable latency, or how long it takes the solution to start transmitting an order after it has received all the necessary information to evaluate whether to send an order. Under repeated tests, the maximum latency metric remained consistent, varying no more than 2 nanoseconds, or within the uncertainty of the test harness itself. These results demonstrate the solid, record-setting performance of the Ultimate Trading Machine.
Facing tremendous pressures to optimize the transaction lifecycle, the financial services industry helps drive innovations in many core technologies. At Penguin Computing, we empower our customers with with open technology solutions that achieve performance requirements while keeping costs low and avoiding vendor lock-in. We are proud to join with our partners to deliver this Ultimate Trading Machine.
About Penguin Computing
Penguin Computing® is the largest, private, North American supplier of Linux®-focused hardware, software, and services for enterprise data center, cloud, and high-performance computing. Leveraging decades of experience, Penguin Computing pioneers the design and engineering of open technology solutions and delivers top-of-class services and support.
As one of a limited number of authorized Open Compute Project vendors, Penguin Computing delivers the benefits of efficiency and density to a broader market with its Tundra® platform for HPC and scale-out data center. Penguin Computing also built and operates the leading, public HPC cloud service, Penguin Computing On-DemandTM.
Penguin Computing has systems installed with more than 2,500 customers in 40 countries across eight major vertical markets.
Solarflare is pioneering server connectivity for neural-class networks. From silicon to firmware to software, Solarflare provides a comprehensive, integrated set of technologies for distributed, ultra-scale, software-defined datacenters.
The Solarflare XtremeScale Architecture is a design framework which includes a comprehensive suite of features for ultra-scale environments: High-bandwidth, ultra-low-latency, ultra-scale connectivity, software defined, secure with hardware firewalls, and instrumented for line-speed telemetry.
Solarflare solutions have earned a sterling reputation in financial services and are used by virtually every major global exchange, commercial bank and hedge fund. This exacting, regulated performance uniquely qualifies our solutions for use in ultra-scale applications in IoT, big data and artificial intelligence where low latency, robust security and insightful telemetrics are critical.
Learn more at www.solarflare.com.
About LDA Technologies
LDA is a group of leading engineers that have been working together in various companies and on numerous projects spanning across multiple industries. A highly experienced and efficient team that formed over 15 years ago, well before the LDA itself, and that has extensive knowledge and years of experience designing products for different industries such as networking, information security, and trading.
Learn more at www.ldatech.com.
About Xilinx Kintex UltraScale FPGAs
For many years, NICs with Kernel Bypass have been used to connect trading application servers to Ethernet networks. Solarflare and LDA achieve breakthrough performance by adding to the processing power of NICs, and by implementing a hyper-efficient TCP stack in an FPGA. Additional processing power is added with Xilinx Kintex UltraScale FPGAs, while an innovative technique called “Delegated TCP send” uniquely delivers lower latency by using several parallel TCP connections. The new class of distributed solutions for TCP was co-developed by combining the LDA Lightspeed TCP IP Core, which can run on multiple Xilinx FPGA-based boards, with the Delegated Sends feature available with Onload™-enabled Solarflare network adapters to manage and maintain TCP connections.
Learn more at www.xilinx.com.
The STAC Benchmark Council and STAC-T0 Benchmark
STAC is a company that coordinates a community called the STAC Benchmark Council. The STAC Benchmark Council consists of over 300 financial institutions and more than 50 vendor organizations whose purpose is to explore technical challenges and solutions in financial services and to develop technology benchmark standards that are useful to financial organizations. User firms include the largest global banks, brokerage houses, exchanges, hedge funds, proprietary trading shops, and other market participants. Vendor firms include innovative manufacturers of processor, server, storage, and network hardware, as well as horizontal-market software providers and vertical specialists.
Learn more at stacresearch.com.