Skip to content

Performance Documentation

Performance analysis, benchmarking, and system optimization guides for ShadowHound.

Overview

This directory tracks performance characteristics, bottleneck analysis, and optimization strategies for the ShadowHound system.

Documents

Key Metrics

LLM Backend Performance

  • Response latency (vLLM vs Ollama vs cloud)
  • Token throughput
  • Memory usage
  • See LLM Integration Hub

Robot Control Performance

  • Command execution latency
  • Skill execution time
  • ROS2 topic rates
  • WebRTC stream quality

System Resources

  • CPU utilization (laptop, Thor, GO2)
  • Memory consumption
  • Network bandwidth
  • Power consumption

Validation

  • [ ] Baseline performance metrics documented
  • [ ] Bottlenecks identified and tracked
  • [ ] Optimization efforts measured and validated

See Also

References