Live Trading

Real-Time Factor Computation

High-frequency market events — matched trades, tick quotes — flow into DolphinDB via stream tables, where vectorized SQL and DLang enable flexible expression of both stateless and stateful sliding-window factors. Complex logic cascades across reactive state and time-series engines for seamless pipeline processing. Multi-instrument, multi-factor parallel execution, incremental computation, and JIT acceleration push real-time factor updates to milliseconds or below — all without compromising development speed.

Featured Tutorials & Best Practices:

• Calculating High-Frequency Factors in Real-Time
• Real-time scenarios include: Real-time Rank of Price Rate of Change](https://docs.dolphindb.com/en/Tutorials/realtime_ROC_rankings.html)

Low-Latency Factor Computation

Swordfish embeds the DolphinDB computation kernel as a native C++ library inside proprietary trading and risk systems — computing factors and risk metrics entirely in-process, with no network round-trips. Type specialization, row-wise execution, and memory optimization bring latency down to tens of microseconds, or single microseconds on high-end hardware — built for tick-level factors, in-path risk checks, and real-time signal generation.

Pricing Engine

DolphinDB loads order book data, tick trades, and external curves — interest rate curves, volatility surfaces — directly into memory, continuously refreshing mid-prices, theoretical prices, and implied volatility via the streaming engine. For options, convertible bonds, and structured products, pricing models and risk factors are computed in-database and pushed to quoting systems as live stream tables — closing the loop from raw market data to pricing to quote decisions in milliseconds, and meeting the latency demands of market making and arbitrage strategies.

Pre-Trade Analysis

Market data, positions, orders, and account information are unified in DolphinDB for real-time in-database computation of duration, convexity, DV01/PVBP, Greeks, and other sensitivity metrics. A configurable rules engine enforces leverage, concentration, liquidity, and drawdown checks with differentiated thresholds per strategy, account, and channel — returning results in milliseconds to intercept or flag risk before orders are sent.

Post-Trade Analysis

Batch replay and analysis of intraday and historical trades — flagging suspected wash trading and volume anomalies by counterparty, instrument, price, and direction. Each fill is benchmarked against multi-source quotes, mid-prices, or VWAP to quantify execution deviation and evaluate broker and algorithm quality. DolphinDB compresses analysis of millions of trades from minutes to milliseconds — turning post-trade review from periodic sampling into systematic, continuous risk control, with findings fed directly back into pre-trade rules and strategy optimization.

Real-Time Monitoring & Alerting

DolphinDB continuously computes key metrics — factor volatility, execution anomalies, PnL, system latency, and queue depth — via streaming pipelines or scheduled tasks, triggering configurable alerts delivered through HTTP, message queues, email, SMS, or enterprise IM.

Monitoring dashboards built on BI tools or internal frontends visualize real-time factors, capital flows, positions, and risk metrics — with interactive drill-down by strategy, account, and instrument.