OSRF's ROS 2 Performance Monitor delivers deep benchmark insights
New ROS 2 benchmark dashboard turns raw data into actionable robotics performance insights...
Ammaar Ahmed, a GSoC 2026 contributor with OSRF, is developing a ROS 2 client library performance monitoring project that connects benchmark execution, result processing, and visualization to make performance results easier to reproduce, compare, and understand. The system runs a reduced rclcpp benchmark matrix, normalizes raw outputs into a consistent format, and launches a Prometheus/Grafana dashboard for interactive analysis. The dashboard compares two benchmark runs at a time, highlights key metrics, and includes a manual explorer for matching exact workloads by topology, middleware, communication mode, payload size, and process layout. It tracks latency, throughput, CPU usage, memory usage, and message reliability where available, while preserving benchmark metadata. The public dashboard currently contains five Jazzy runs, five Lyrical runs, and one median summary for each distribution, with every benchmark running for 60 seconds. The project is still in progress, but the main workflow already works from benchmark execution to a public dashboard.
- First public dashboard compares ROS 2 Jazzy vs Lyrical with 5 benchmark runs each (60-second tests)
- Automated pipeline converts raw rclcpp benchmarks → JSONL → Grafana dashboard with Prometheus backend
- Tracks 6 key metrics (latency, throughput, CPU, memory, reliability, message size) plus full configuration metadata
Why It Matters
Gives robotics engineers empirical data to optimize ROS 2 deployments and make informed middleware choices