Skip to main content

How to Store and Manage ROS Data

· 29 min read
Anthony Cavin
Data Scientist - ML/AI, Python, TypeScript

ROS 2 Data Storage Tutorial

In this tutorial, we will create a custom ROS 2 Humble package called rosbag2reduct that records incoming ROS 2 topics into MCAP bag files on a Raspberry Pi and automatically uploads those files to a ReductStore instance with metadata labels. We'll walk through setting up ROS 2 Humble on the Pi, interfacing a USB camera using the v4l2_camera driver, deploying a lightweight YOLOv5 (nano) object detection node (using ONNX Runtime) to produce detection metadata, and implementing the rosbag2reduct node to capture data and offload it. We will also cover installing ReductStore on the Pi, configuring replication of labeled data to a central storage on your laptop (using label-based filters via the web console). This end-to-end guide is structured with clear steps, code examples, and configuration snippets to help you build and deploy the system.

ReductStore vs. MinIO: Beyond Benchmarks

· 7 min read
Anthony Cavin
Data Scientist - ML/AI, Python, TypeScript

MinIO vs ReductStore

As data-driven applications evolve, the need for efficient storage solutions continues to grow. ReductStore and MinIO are two powerful solutions designed to handle massive amounts of unstructured data, but they serve different purposes.

While ReductStore is optimized for time-series object storage with a focus on unstructured data such as sensor logs, images, and machine-generated data for robotics and IIoT, MinIO is a high-performance object storage system built for scalable, cloud-native applications with a focus on S3 compatibility and enterprise-wide storage needs.

In this article, we'll explore the differences between ReductStore and MinIO, examine where each excels, and discuss how they can be used together to build a more comprehensive data storage solution.

Building a Data Acquisition System for Manufacturing

· 13 min read
Alexey Timin
Software Engineer - Database, Rust, C++

Large manufacturing plants generate vast amounts of data from machines and sensors. This data is valuable for monitoring machine health, predicting failures, and optimizing production. It also serves as a foundation for building industrial AI models for predictive maintenance, quality control, and process optimization.

A Data Acquisition (DAQ) system collects this data, processes it, and stores it for further analysis. It typically consists of edge devices that gather real-time data, central servers or cloud storage for retention, and software that enables analytics and AI-driven insights.

DAQ System based on ReductStore

An example of a 3 tier DAQ system based on ReductStore.

Traditional automation solutions like SCADA and historians are complex, expensive, and not optimized for modern cloud-based AI applications. They often limit access to data, making it difficult for engineers and data scientists to develop machine learning models and gain actionable insights.

In this article, we’ll explore the challenges of building a modern DAQ system for manufacturing and how ReductStore can simplify the process and support ELT (Extract, Load, Transform) workflows for advanced analytics and AI applications.