GPU Metrics (Upcoming)

We're actively working to add GPU metrics support in the next updates. Nowadays GPU plays a crucial role in Artificial Intelligence (AI) and Machine Learning (ML) workflows. We believe that the current profiling systems does not integrate seamlessly in the current production systems that are constantly demanding resources. Optimization is quick becoming an essential feature in order to avoid resources and budget wasting and so we're constantly evolving to provide a better optimized platform.

See also

• Integrated Metrics - the live eBPF/gRPC metrics and the incoming OpenTelemetry suite
• Architecture Overview - the four-stage pipeline (kernel → maps → agent → consumption)
• Development Goals & Discussions - the "GPU observability" roadmap track