The cryptocurrency market’s meteoric rise has ushered in a new era of trading opportunities. The proliferation of exchanges and cryptocurrencies, round-the-clock trading, and high volatility create a fertile ground for arbitrage, especially appealing to quantitative trading firms.

To capitalize on these market dynamics, sophisticated investment research and trading systems are essential. However, the varying technical capabilities among crypto trading institutions highlight the need for an integrated solution that can streamline strategy development and execution in quantitative trading. Consequently, the need for a comprehensive data management platform arose. As a powerful real-time platform for analytics and stream processing, DolphinDB is dedicated to providing one-stop cutting-edge analytics solutions for global quantitative traders.

Recently, DolphinDB released the whitepaper “Cryptocurrency Solutions” which introduces comprehensive solutions for cryptocurrency trading. Leveraging extensive experience serving financial institutions, DolphinDB crafted a tailored approach that addresses key challenges for quantitative crypto trading in data access and storage, strategy development, and real-time computation. By implementing DolphinDB’s solution, traders and developers can unlock unprecedented performance, dramatically boost development efficiency, and significantly reduce operational costs.

Download our whitepaper for a new crypto management experience: Cryptocurrency Solutions — Dolphindb

Chapter 1 Overview

1.1 Data Processing Workflow

The processing workflow for cryptocurrency market data consists of three components: data sources, computation layer, and consumption layer.

Figure 1–1 Data Processing Workflow

Data Source

DolphinDB efficiently processes both real-time market streams and historical data.

For real-time market streams (such as order information, tick-by-tick trade and quote data), DolphinDB offers the following tools for data ingestion:

• Multiple APIs (e.g., Python, Java) supporting direct data feeds from exchanges.

• Communication plugins (e.g., httpClient, WebSocket) for accessing cryptocurrency APIs.

• Message middleware plugins (e.g., Kafka, ZeroMQ) for data subscription via configured message queue topics.

Additionally, DolphinDB offers a replay feature that simulates real-time streams using historical data for development and debugging purposes. This can also be utilized for stress testing to assess stream computing code stability.

Computation Layer

DolphinDB provides robust computational capabilities for both stream and batch processing:

• Real-time stream computing: Built-in streaming engines support time series calculations (e.g., minute-level OHLC), indicator calculations (e.g., MACD, KDJ), cross-sectional calculations, stream joins, etc.

• Batch processing: Distributed batch processing computations are supported through a comprehensive, user-friendly programming language and extensive built-in functions, enabling efficient processing and analysis of large-scale data.

Consumption Layer

The computing results can be saved to in-memory tables or to DFS tables on disk. Downstream applications can consume the results through DolphinDB APIs or message plugins.

1.2 Unified Batch and Stream Processing

DolphinDB implements unified batch and stream processing. Factor expressions developed in research environments can be directly applied to real-time processing in production, ensuring consistent results of development and live trading. The unified framework significantly reduces implementation time compared to traditional codebases that use separate languages for development (e.g., Python) and live trading (e.g., C++).

Figure 1–2 Unified Batch and Stream Processing

1.3 Key Features for Backtesting

DolphinDB offers a suite of advanced features designed to streamline the backtesting process for crypto strategy development. The backtesting solution comprises three key components:

• Historical Data Replay: This feature replays historical data as real-time streams, enabling developers to observe how their strategies would perform in live trading.

• Order Matching Simulator: The order matching simulator is a critical component for validating strategy effectiveness. DolphinDB provides a high-precision matching simulator that aligns with live trading conditions, adhering to the “price-time priority” principle and supporting various configurations of matching mechanisms.

• Backtesting Framework: DolphinDB backtesting framework supports strategy development and performance evaluation. It accommodates a wide range of indicators, models, and machine learning methods, while supporting backtesting based on order book or OHLC data. The system provides detailed information on strategy returns, positions, and trade details.

The above introduces an overview of DolphinDB framework for cryptocurrency solutions, covering the data processing flow, unified batch and stream processing, and key features for backtesting. In the next chapter, we will introduce market data access, integration and storage. Stay tuned for the latest updates!

Download Whitepaper Here: Cryptocurrency Solutions — Dolphindb

Email us for more information or to schedule a demo: sales@dolphindb.com

Thanks for your reading! To keep up with our latest news, please follow our Twitter @DolphinDB_Inc and Linkedin. You can also join our Slack to chat with the author! 😉