profiling-hostbound
CommunityDeep analysis of host and NPU profiling data for performance bottlenecks.
Authordongg622
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill enables comprehensive analysis of profiling and trace data to identify performance issues related to host processes and NPU operations.
Core Features & Use Cases
- Performance Bottleneck Identification: Detect host-side process issues such as high context switches, CPU migrations, and long interrupt handling durations.
- NPU Performance Analysis: Identify slow or imbalanced NPUs by analyzing calculation and communication times across multiple hardware devices.
- Use Case: When system performance slows down during large model inference, this Skill helps pinpoint whether the problem stems from host scheduling issues or NPU bottlenecks.
Quick Start
Analyze the trace file 'trace.log' to generate performance analysis reports and visualizations.
Dependency Matrix
Required Modules
None requiredComponents
scriptsreferences
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: profiling-hostbound Download link: https://github.com/dongg622/china-ai-chip-skill/archive/main.zip#profiling-hostbound Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.