Fontes Fintech

Why CPU Affinity and Core Pinning Matter in Low-Latency Trading SystemsTrading Engineering — Deep Dive #3In low-latency trading systems, CPU utilization alone does not tell the whole story.It also matters which thread runs on which CPU core, how often threads move between cores, and how efficiently each thread can reuse the data already present in the CPU cache.This is where CPU Affinity and Cor..

저지연 멀티스레드 시스템에서 False Sharing이 중요한 이유Trading Engineering — Deep Dive #2저지연 트레이딩 시스템에서는 CPU Cache를 이해하는 것만으로 충분하지 않습니다.데이터가 Cache에 올라와 있더라도 서로 다른 스레드가 데이터를 어떻게 배치하고 접근하는가에 따라 성능이 크게 달라질 수 있습니다.그 대표적인 문제가 False Sharing입니다.False Sharing은 여러 스레드가 서로 다른 데이터를 사용하고 있음에도 불구하고, 그 데이터가 같은 Cache Line에 위치하면서 불필요한 Cache Line 이동과 동기화 비용이 발생하는 현상입니다.특히 다음과 같은 환경에서 문제가 될 수 있습니다.멀티스레드 Trading SystemMarket Data P..

How Does an Algorithmic Trading System Process an Order?Trading Engineering Series — 06Algorithmic trading systems are designed to make trading decisions and execute orders automatically based on predefined strategies.However, an algorithmic trading system is not simply a program that generates buy or sell orders.From receiving market data to generating a trading signal, performing pre-trade ris..

Five Factors That Determine Latency in Trading SystemsTrading Engineering Series — 05 When developing a Low Latency Trading System, the first things that often come to mind are CPU performance and network speed.However, trading system latency is not determined by a single factor.From the moment market data enters the system to the moment an order reaches the exchange, many small operations contr..