| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | ||||
| 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| 11 | 12 | 13 | 14 | 15 | 16 | 17 |
| 18 | 19 | 20 | 21 | 22 | 23 | 24 |
| 25 | 26 | 27 | 28 | 29 | 30 | 31 |
- fontesfintech
- kernel bypass
- ultra low latency
- hft programming
- Korea HFT
- algo trading
- speed trading
- dma development
- mmlp
- trading engineering
- hft개발
- hft tuning
- 주문 오더북
- dma fep
- dma hft
- Korea dma
- system trading
- 자동매매
- 저지연시스템
- korea inbound
- fep개발
- HFT system
- 트레이딩 엔지니어링
- DMA개발
- 시스템 트레이딩
- 프랍데스크
- 주문fep
- high frequency trading
- low latency
- nic offload
- Today
- Total
목록2026/10/05 (3)
Fontes Fintech
Memory Ordering, Atomic Operations and Memory Barriers in Low-Latency Trading SystemsTrading Engineering — Deep Dive #10IntroductionWhen we talk about performance in low-latency trading systems, we eventually arrive at a fundamental question:How do multiple CPU cores exchange and observe data correctly while running concurrently? In a simple program, we may imagine:Thread A ↓Memory Write ↓Th..
Inside a Low-Latency Trading Network: UDP Market Data and TCP Order FlowTrading Engineering — Deep Dive #9IntroductionWhen discussing networking in low-latency trading systems, simply comparing TCP and UDP is not enough.In a real trading system, market data and order flow follow different network paths and have different engineering requirements.A simplified exchange-connected trading architectu..
저지연 트레이딩 시스템에서 Memory Ordering과 Atomic Operation이 중요한 이유Trading Engineering — Deep Dive #10Introduction저지연 트레이딩 시스템의 성능을 이야기하다 보면 어느 순간 이런 질문에 도달하게 됩니다.여러 CPU Core에서 동시에 실행되는 Thread들은 서로의 데이터를 어떻게 정확하게 볼 수 있을까?단순한 프로그램에서는 다음과 같이 생각하기 쉽습니다.Thread A ↓Memory Write ↓Thread B ↓Memory Read 하지만 실제 멀티코어 CPU에서는 훨씬 복잡합니다.각 CPU Core는 자신만의 Cache를 가지고 있고, Store Buffer와 Load/Store 처리 구조를 사용하며, Compiler ..
