🚀 Listed in the CNCF Landscape and eBPF Foundation Emerging Project. If you like this project, give it a star on GitHub! ❤️

Kernel Insight

Low Overhead
Zero Instruction
Intelligent Adaptive Backoff

Instant Observability

Context Events
Full-stack Deep Insight
Key-Path Instrumentation

Continuous Profiling

Multi-Language
Memory, CPU, Lock
Driving Business Innovation

Autotracing

Tracing Engine
Automated Snapshots
Resolution for System Jitter

Kernel-wide Insight

Leveraging BPF to deliver full-stack observability and deep performance insights into memory, scheduling, networking, and block I/O, with 1% overhead.

Instant Observability

Event-driven context snapshot with instrumentation across kernel slow paths. Automatic reports for system-wide events, including scheduling, networking, and interrupts.

AutoTracing

Pinpoint performance degradation in cloud-native environments with tracing-based intelligence. Automated tracing for CPU idle drops, I/O latency, and Loadavg spikes.

Continuous Profiling

Continuous profiling of OS kernel and multi-language (e.g. Java, Python, Go, C/C++.) applications across CPU, memory, I/O, and Lock. Driving business innovation.

Heterogeneous Hardware

Encompassing core system components such as CPUs, memory, PCIe interconnects, network adapters, and storage, alongside specialized AI accelerators including GPUs and NPUs.

Ecosystem Integration

Integrate open‑source observability stacks. Support bare‑metal and cloud‑native deployments. Aware of K8s containers / labels / annotations. Support mainstream Linux distributions.

Best Practice

Cloud Native
Kubernetes, Distributed System

Provide observability metrics, kernel events, tracing, and profiling for containers, physical machines, and controllers.
AI Computing
Large model training and inference

Extend GPU fault detection, delivering metrics and events for CPU, Memory, PCIe, HCA, and other components.
Agent Sandbox
General-purpose AI Agent Sandbox

Designed for the general-purpose sandbox runtime and serverless computing, e.g. Kata, Firecracker, Cloud Hypervisor microVMs.
Cloud Computing
Integrated Linux Operating System

Support for Huawei EulerOS, Alibaba Anolis OS, Tencent TencentOS, and other mainstream operating systems.
Infrastructure Services
Data Center, Bare Metal

Apply to infrastructure services, including storage, big data, message queues, microservices, and other systems.
Chaos Engineering
System Reliability and Stability

Support chaos engineering, fault injection, and data center disaster recovery, enabling users to understand application.