Real-Time Stock Market Data Processing with Big Data Technologies

Introduction

This project focuses on the real-time processing and analysis of stock market data using Big Data technologies. It aims to provide deep insights and timely information about market trends for effective investment strategies and risk management.

Technologies Used

• Docker: For containerization and simulation of distributed environments.
• Apache Kafka: For managing and distributing real-time data streams.
• Apache Spark & Spark Streaming: For real-time data processing and analysis.
• Apache Spark SQL & MLlib: For data querying and machine learning tasks.
• Hadoop HDFS: For storage and batch processing of large datasets.
• InfluxDB & Grafana: For time-series data storage and visualization.
• MongoDB: For NoSQL data storage.

System Architecture

The system uses a lambda architecture model, integrating both batch and stream processing for comprehensive data analysis. The architecture details can be found in the project report.

Installation and Setup

Clone project source code:

git clone https://github.com/datphamvn/big-data-stock-analysis.git

Set up and change config (if necessary) at config-hadoop and .env

Usage

To initialize and execute components in the Docker environment, use commands:

docker-compose --env-file ../.\.env up -d --remove-orphans --build

Launch producer

cd producer
python producer.py

The interface visualized through Grafana is displayed at: http://localhost:3000/

Contributors

• 20210158 Phạm Vũ Tuấn Đạt
• 20215472 Trần Cao Sơn
• 20215401 Trần Quang Khải
• 20210329 Hoàng Nghĩa Hiệp
• 20210784 Nguyễn Phúc Thắng