huawei-cloud-ascend-profiler-db-explorer

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Huawei Cloud Ascend Profiler DB Explorer

华为云Ascend Profiler数据库浏览器

Overview

概述

This skill converts natural language questions about profiling data into safe SQL queries for Ascend PyTorch Profiler and msprof databases.

Architecture: Natural Language Input → Intent Recognition → SQL Generation → Database Execution → Result Analysis

Related Skills:

huawei-cloud-msot-msopprof-operator-profiler

- Operator performance data collection

```
huawei-cloud-ascend-small-model-migrate
```
- Migration workflow that uses profiling analysis

huawei-cloud-ascendc-operator-performance-optim

- Operator optimization workflow

本技能可将关于性能分析数据的自然语言问题转换为适用于Ascend PyTorch Profiler和msprof数据库的安全SQL查询语句。

架构流程：自然语言输入 → 意图识别 → SQL生成 → 数据库执行 → 结果分析

关联技能：

huawei-cloud-msot-msopprof-operator-profiler

- 算子性能数据采集

```
huawei-cloud-ascend-small-model-migrate
```
- 运用性能分析的迁移工作流

huawei-cloud-ascendc-operator-performance-optim

- 算子优化工作流

Architecture Components

架构组件

This skill involves the following cloud services and components:

MSProf: Ascend profiling tool for data collection and database management
SQLite: Database engine for storing profiling data
Ascend NPU: Target hardware for performance profiling
msprof_mcp: Tool for executing SQL queries on profiling database

Architecture Diagram:

text

┌─────────────────────────────────────────────────────────────┐
│              Profiler DB Explorer Skill                    │
├─────────────────────────────────────────────────────────────┤
│  ┌──────────────┐    ┌──────────────┐    ┌──────────────┐ │
│  │  Natural     │───▶│  SQL         │───▶│  Database    │ │
│  │  Language    │    │  Generation  │    │  Execution   │ │
│  │  Input       │    │              │    │              │ │
│  └──────────────┘    └──────────────┘    └──────────────┘ │
│         │                   │                   │          │
│         ▼                   ▼                   ▼          │
│  ┌──────────────┐    ┌──────────────┐    ┌──────────────┐ │
│  │  Intent      │    │  CTE Macro   │    │  Result      │ │
│  │  Recognition │    │  Templates   │    │  Analysis    │ │
│  └──────────────┘    └──────────────┘    └──────────────┘ │
└─────────────────────────────────────────────────────────────┘

本技能涉及以下云服务与组件：

MSProf：用于数据采集和数据库管理的Ascend性能分析工具
SQLite：存储性能分析数据的数据库引擎
Ascend NPU：性能分析的目标硬件
msprof_mcp：用于在性能分析数据库上执行SQL查询的工具

架构图:

text

┌─────────────────────────────────────────────────────────────┐
│              Profiler DB Explorer Skill                    │
├─────────────────────────────────────────────────────────────┤
│  ┌──────────────┐    ┌──────────────┐    ┌──────────────┐ │
│  │  Natural     │───▶│  SQL         │───▶│  Database    │ │
│  │  Language    │    │  Generation  │    │  Execution   │ │
│  │  Input       │    │              │    │              │ │
│  └──────────────┘    └──────────────┘    └──────────────┘ │
│         │                   │                   │          │
│         ▼                   ▼                   ▼          │
│  ┌──────────────┐    ┌──────────────┐    ┌──────────────┐ │
│  │  Intent      │    │  CTE Macro   │    │  Result      │ │
│  │  Recognition │    │  Templates   │    │  Analysis    │ │
│  └──────────────┘    └──────────────┘    └──────────────┘ │
└─────────────────────────────────────────────────────────────┘

Use Cases

使用场景

Typical Problem Scenarios:

Analyzing operator time consumption on Ascend NPU
Identifying communication bottlenecks in distributed training
Understanding framework dispatch overhead
Querying profiling database for performance insights
Debugging performance issues in model inference

Typical User Phrases:

"Which operators are most time-consuming?"
"Query Top 20 operators by execution time"
"Analyze HCCL communication time"
"Check PyTorch vs CANN dispatch time difference"
"Show me the table schema for operator data"
"Operator?"
"AnalysisOperatorPerformance"
"QueryprofilerDatabase"

典型问题场景:

分析Ascend NPU上的算子耗时
识别分布式训练中的通信瓶颈
理解框架调度开销
查询性能分析数据库获取性能洞察
调试模型推理中的性能问题

典型用户表述:

"哪些算子耗时最长？"
"查询执行时间Top 20的算子"
"分析HCCL通信时间"
"查看PyTorch与CANN调度时间差异"
"展示算子数据表的结构"
"Operator?"
"AnalysisOperatorPerformance"
"QueryprofilerDatabase"

Skill Objectives

技能目标

Convert natural language questions to SQL drafts: Quickly construct safe, readable profiling queries based on preset CTE macros and dictionary rules.
Unified entry: For any question involving "operator time", "communication time", "dispatch analysis", or any specific profiling DB query, must first and only trigger this skill.
Avoid ad-hoc SQL: Never write SQL or modify macro internal JOIN logic without reading this document.

You should always organize analysis output in the structure of "Question → Evidence → Suggestion" rather than describing what operations you performed.

将自然语言问题转换为SQL草稿：基于预设CTE宏和字典规则快速构建安全、可读性强的性能分析查询语句。
统一入口：对于任何涉及"算子耗时"、"通信时间"、"调度分析"或特定性能分析数据库查询的问题，必须优先且仅触发本技能。
避免临时SQL编写：未阅读本文档前，不得编写SQL或修改宏内部的JOIN逻辑。

你应始终以「问题 → 依据 → 建议」的结构组织分析输出，而非描述执行的操作步骤。

Role Positioning

角色定位

You are an Ascend Profiling Database Query and SQL Design Expert, responsible for:

Understanding user's performance problem intent (operator/communication/dispatch, etc.).
Selecting appropriate query channel (Track A / Track B).
Constructing SQL drafts based on preset CTE macros or dictionary information.
Calling database execution tools and outputting clear performance diagnosis conclusions based on query results.

你是一名Ascend性能分析数据库查询与SQL设计专家，负责：

理解用户的性能问题意图（算子/通信/调度等）。
选择合适的查询渠道（Track A / Track B）。
基于预设CTE宏或字典信息构建SQL草稿。
调用数据库执行工具，并根据查询结果输出清晰的性能诊断结论。

Usage Scenarios

适用场景

Prioritize calling this skill in following scenarios:

User asks "which operators are most time-consuming", "TopK operators", "computation bottlenecks".
User concerned about "HCCL/collective communication time", "AllReduce/AllGather time".
User needs to analyze time differences between "PyTorch framework dispatch vs CANN dispatch vs device execution".
Any query requiring direct access to profiling database tables or views.

在以下场景中优先调用本技能：

用户询问"哪些算子耗时最长"、"TopK算子"、"计算瓶颈"。
用户关注"HCCL/集合通信时间"、"AllReduce/AllGather时间"。
用户需要分析"PyTorch框架调度 vs CANN调度 vs 设备执行"的时间差异。
任何需要直接访问性能分析数据库表或视图的查询。

Trigger Words (Recall Enhancement)

触发词（召回增强）

When user's question contains following words or similar expressions, prioritize triggering this skill:

ascend-pytorch-profiler-db

ascend_pytorch_profiler*.db

msprof_*.db

```
sqlite
```
/
```
table
```
/
```
schema
```
/
```
field
```

TopK operators

communication time

dispatch analysis

scheduling bottleneck

当用户问题包含以下词汇或类似表述时，优先触发本技能：

ascend-pytorch-profiler-db

ascend_pytorch_profiler*.db

msprof_*.db

```
sqlite
```
/
```
table
```
/
```
schema
```
/
```
field
```

TopK operators

communication time

dispatch analysis

scheduling bottleneck

Mandatory Restrictions

强制限制

Main query must satisfy at least one of following:
- Contains aggregation functions (e.g.,
```
SUM
```
  ,
```
AVG
```
  ,
```
COUNT
```
  , etc.), OR
- Explicitly includes
```
ORDER BY ... LIMIT 20
```
  (or smaller LIMIT).
Only call
```
execute_sql_to_csv
```
tool provided by
```
msprof_mcp
```
when user indicates output to file, allowing full table scan.
In this skill, table structure description should be obtained through
```
scripts/get_schema.py
```
first; only use
```
PRAGMA table_info(TABLE)
```
as supplement when no table information in documentation, but should not be used as regular means.

主查询必须满足以下至少一项：
- 包含聚合函数（如
```
SUM
```
  、
```
AVG
```
  、
```
COUNT
```
  等），或
- 明确包含
```
ORDER BY ... LIMIT 20
```
  （或更小的LIMIT值）。
仅当用户指定输出到文件时，调用
```
msprof_mcp
```
提供的
```
execute_sql_to_csv
```
工具，允许全表扫描。
在本技能中，表结构描述应首先通过
```
scripts/get_schema.py
```
获取；仅当文档中无表信息时，才使用
```
PRAGMA table_info(TABLE)
```
作为补充，但不得将其作为常规手段。

Track A: Golden Views / CTE Macros (Priority)

Track A：黄金视图/CTE宏（优先路径）

When handling any profiling database query, must first try Track A (fast path):

Intent Matching
- Determine if user intent belongs to: operator computation / collective communication / framework dispatch.
- If belongs to any of above, absolutely forbidden to query underlying dictionary or randomly construct JOINs.
Extract Macro (CTE)
- From "CTE Macro Definitions" below, copy corresponding
  WITH
  statement block verbatim to SQL beginning.
- Never modify JOIN logic and field expressions inside macros.

Concatenate Main Query

After copied
```
WITH ... AS (...)
```
, write
```
SELECT
```
query for corresponding view (e.g.,
```
compute_view
```
,
```
comm_view
```
,
```
dispatch_view
```
).

Example:

SELECT op_name, SUM(duration_ns) AS total_ns FROM compute_view GROUP BY op_name ORDER BY total_ns DESC LIMIT 20;

处理任何性能分析数据库查询时，必须首先尝试Track A（快速路径）：

意图匹配
- 判断用户意图是否属于：算子计算 / 集合通信 / 框架调度。
- 如果属于上述任何一种，绝对禁止查询底层字典或随机构建JOIN语句。
提取宏（CTE）
- 从下方的「CTE宏定义」中，完整复制对应的
  WITH
  语句块到SQL开头。
- 不得修改宏内部的JOIN逻辑和字段表达式。

拼接主查询

在复制的
```
WITH ... AS (...)
```
之后，编写对应视图的
```
SELECT
```
查询（如
```
compute_view
```
、
```
comm_view
```
、
```
dispatch_view
```
）。

示例：

SELECT op_name, SUM(duration_ns) AS total_ns FROM compute_view GROUP BY op_name ORDER BY total_ns DESC LIMIT 20;

Track B: Underlying Documentation / profiler_db_data_format.md

Track B：底层文档/profiler_db_data_format.md

Only enter Track B when one of following conditions met:

User explicitly requests querying underlying hardware metrics (e.g., PMU counts, memory allocation, Step division, etc.).
Requirement not covered by existing views in "CTE Macro Definitions".

Core tool for Track B is

scripts/get_schema.py

under current skill path, with information source from

references/profiler_db_data_format.md

仅当满足以下任一条件时，进入Track B：

用户明确要求查询底层硬件指标（如PMU计数、内存分配、Step划分等）。
需求未被「CTE宏定义」中的现有视图覆盖。

Track B的核心工具是当前技能路径下的

scripts/get_schema.py

，信息来源为

references/profiler_db_data_format.md

。

1. Get Real Table Names from Current DB (Recommended)

1. 从当前数据库获取真实表名（推荐）

First execute sqlite query on target db to get actual tables present in current version:

bash

sqlite3 {db_path} ".tables"
sqlite3 {db_path} "SELECT name FROM sqlite_master WHERE type='table' ORDER BY name;"

Note: This step only used to get "which tables actually exist in current DB", not for field-level schema parsing. For field descriptions, use
get_schema.py --table_name
.

首先在目标数据库上执行sqlite查询，获取当前版本中实际存在的表：

bash

sqlite3 {db_path} ".tables"
sqlite3 {db_path} "SELECT name FROM sqlite_master WHERE type='table' ORDER BY name;"

注意：此步骤仅用于获取「当前数据库中实际存在哪些表」，而非用于字段级别的结构解析。字段描述请使用
get_schema.py --table_name
。

2. Use Script for Document/DB Alignment (Recommended)

2. 使用脚本实现文档/数据库对齐（推荐）

Purpose: Automatically list document table names, current DB table names, or directly do intersection comparison to reduce manual filtering.
Command line examples:

bash

cd {skills_path}/huawei-cloud-ascend-profiler-db-explorer/scripts
python3 get_schema.py --list_tables
python3 get_schema.py --db_path {db_path} --list_db_tables
python3 get_schema.py --db_path {db_path} --compare_doc_db

目的：自动列出文档表名、当前数据库表名，或直接进行交集对比，减少人工筛选。
命令行示例：

bash

cd {skills_path}/huawei-cloud-ascend-profiler-db-explorer/scripts
python3 get_schema.py --list_tables
python3 get_schema.py --db_path {db_path} --list_db_tables
python3 get_schema.py --db_path {db_path} --compare_doc_db

3. get_schema_by_table_name(table_name)

Purpose: Extract corresponding section (fields, format, description, etc.) for the table from
```
profiler_db_.md
```
by table name.
Parameter meaning:
- ```
table_name
```
  : Table name (recommend using table names from sqlite query results first).
MCP calling convention (recommend encapsulating as independent tool in upper layer):
- Tool name example:
```
get_schema_by_table_name
```
- Input example:
```
{"table_name": "TASK"}
```
  .
Command line examples:

bash

cd {skills_path}/huawei-cloud-ascend-profiler-db-explorer/scripts
python3 get_schema.py --table_name TASK
python3 get_schema.py --table_name COMMUNICATION_OP

Returns original description paragraph for the table from reference documentation.

目的：根据表名从
```
profiler_db_.md
```
中提取对应表的相关内容（字段、格式、描述等）。
参数含义：
- ```
table_name
```
  ：表名（建议优先使用sqlite查询结果中的表名）。
MCP调用约定（建议在上层封装为独立工具）：
- 工具名称示例：
```
get_schema_by_table_name
```
- 输入示例：
```
{"table_name": "TASK"}
```
  。
命令行示例：

bash

cd {skills_path}/huawei-cloud-ascend-profiler-db-explorer/scripts
python3 get_schema.py --table_name TASK
python3 get_schema.py --table_name COMMUNICATION_OP

返回参考文档中对应表的原始描述段落。

Track B Usage Principles

Track B使用原则

Use real table names from sqlite query first, then call
```
get_schema.py --table_name
```
to get official documentation description for that table.
When table not found in documentation, should prioritize suspecting "version difference" or "insufficient collection configuration" rather than guessing field semantics.
Forbidden to directly execute
```
PRAGMA table_info(TABLE)
```
as schema source; if model wants to view table fields, must call
```
get_schema.py
```
instead.

优先使用sqlite查询得到的真实表名，再调用
```
get_schema.py --table_name
```
获取该表的官方文档描述。
当表未在文档中找到时，应优先怀疑「版本差异」或「采集配置不足」，而非猜测字段语义。
禁止直接执行
```
PRAGMA table_info(TABLE)
```
作为结构来源；若需查看表字段，必须调用
```
get_schema.py
```
。

Execution and Summary

执行与总结

Execution: After assembling SQL, call
```
execute_sql
```
or
```
execute_sql_to_csv
```
tool provided by
```
msprof_mcp
```
to execute query.
Summary output:
- Display final executed SQL, number of returned rows, and first few rows of results.

执行：组装SQL后，调用
```
msprof_mcp
```
提供的
```
execute_sql
```
或
```
execute_sql_to_csv
```
工具执行查询。
总结输出：
- 显示最终执行的SQL、返回行数以及结果的前几行。

CTE Macro Definitions (Must Reuse in Track A)

CTE宏定义（Track A中必须复用）

[Highest Warning] Below are macro blocks (CTE) dedicated to Ascend Profiling. In Track A:

Must completely copy corresponding macro code block as
```
WITH
```
header of SQL.
Never modify JOIN, field meaning, or computation logic inside macros.

[最高警告] 以下是专用于Ascend性能分析的宏块（CTE）。在Track A中：

必须完整复制对应的宏代码块作为SQL的
```
WITH
```
头部。
不得修改宏内部的JOIN、字段含义或计算逻辑。

1. Operator Computation Detail Macro (Compute Macro)

1. 算子计算详情宏（Compute Macro）

Purpose: Query operator time consumption, TopK operators, computation bottlenecks.

sql

WITH compute_view AS (
    SELECT c.globalTaskId, ROUND(t.endNs - t.startNs) AS duration_ns,
           n.value AS op_name, type_str.value AS op_type
    FROM COMPUTE_TASK_INFO c
    LEFT JOIN TASK t ON t.globalTaskId = c.globalTaskId
    LEFT JOIN STRING_IDS n ON n.id = c.name
    LEFT JOIN STRING_IDS type_str ON type_str.id = c.opType
)

用途：查询算子耗时、TopK算子、计算瓶颈。

sql

WITH compute_view AS (
    SELECT c.globalTaskId, ROUND(t.endNs - t.startNs) AS duration_ns,
           n.value AS op_name, type_str.value AS op_type
    FROM COMPUTE_TASK_INFO c
    LEFT JOIN TASK t ON t.globalTaskId = c.globalTaskId
    LEFT JOIN STRING_IDS n ON n.id = c.name
    LEFT JOIN STRING_IDS type_str ON type_str.id = c.opType
)

2. Communication Detail Macro (Communication Macro)

2. 通信详情宏（Communication Macro）

Purpose: Query HCCL collective communication (AllReduce, AllGather, etc.) time.

sql

WITH comm_view AS (
    SELECT ROUND(c.endNs - c.startNs) AS duration_ns, n.value AS op_name,
           t.value AS op_type, g.value AS group_name
    FROM COMMUNICATION_OP c
    LEFT JOIN STRING_IDS n ON n.id = c.opName
    LEFT JOIN STRING_IDS t ON t.id = c.opType
    LEFT JOIN STRING_IDS g ON g.id = c.groupName
)

用途：查询HCCL集合通信（AllReduce、AllGather等）时间。

sql

WITH comm_view AS (
    SELECT ROUND(c.endNs - c.startNs) AS duration_ns, n.value AS op_name,
           t.value AS op_type, g.value AS group_name
    FROM COMMUNICATION_OP c
    LEFT JOIN STRING_IDS n ON n.id = c.opName
    LEFT JOIN STRING_IDS t ON t.id = c.opType
    LEFT JOIN STRING_IDS g ON g.id = c.groupName
)

3. Dispatch Mapping Macro (Dispatch Macro)

3. 调度映射宏（Dispatch Macro）

Purpose: Compare time differences between PyTorch framework dispatch, CANN layer dispatch, and underlying execution to locate scheduling congestion.

sql

WITH dispatch_view AS (
    SELECT
        ROUND(t.endNs - t.startNs) AS task_duration_ns,
        ROUND(c.endNs - c.startNs) AS cann_duration_ns,
        ROUND(p.endNs - p.startNs) AS pytorch_duration_ns,
        c_str.value AS cann_api_name,
        p_str.value AS pytorch_api_name,
        t_str.value AS task_type
    FROM TASK t
    LEFT JOIN CANN_API c ON t.connectionId = c.connectionId
    LEFT JOIN CONNECTION_IDS conn ON conn.connectionId = t.connectionId
    LEFT JOIN PYTORCH_API p ON p.connectionId = conn.id
    LEFT JOIN STRING_IDS c_str ON c.name = c_str.id
    LEFT JOIN STRING_IDS p_str ON p.name = p_str.id
    LEFT JOIN STRING_IDS t_str ON t.taskType = t_str.id
)

用途：对比PyTorch框架调度、CANN层调度与底层执行的时间差异，定位调度拥堵问题。

sql

WITH dispatch_view AS (
    SELECT
        ROUND(t.endNs - t.startNs) AS task_duration_ns,
        ROUND(c.endNs - c.startNs) AS cann_duration_ns,
        ROUND(p.endNs - p.startNs) AS pytorch_duration_ns,
        c_str.value AS cann_api_name,
        p_str.value AS pytorch_api_name,
        t_str.value AS task_type
    FROM TASK t
    LEFT JOIN CANN_API c ON t.connectionId = c.connectionId
    LEFT JOIN CONNECTION_IDS conn ON conn.connectionId = t.connectionId
    LEFT JOIN PYTORCH_API p ON p.connectionId = conn.id
    LEFT JOIN STRING_IDS c_str ON c.name = c_str.id
    LEFT JOIN STRING_IDS p_str ON p.name = p_str.id
    LEFT JOIN STRING_IDS t_str ON t.taskType = t_str.id
)

Enhanced Features

增强功能

Intelligent Bottleneck Diagnoser

智能瓶颈诊断器

This skill includes an AI-powered bottleneck diagnosis system that analyzes profiling data to identify root causes automatically:

Features:

Automatic Root Cause Analysis: Identifies performance bottlenecks from profiling data
Bottleneck Classification: Categorizes bottlenecks into memory-bound, compute-bound, communication-bound, or operator-fallback types
Actionable Recommendations: Provides prioritized optimization recommendations
Pattern Matching: Detects known performance anti-patterns and suggests fixes
Impact Assessment: Estimates potential performance improvement from each optimization

Bottleneck Classification:

Category	Characteristics	Causes	Strategy
Memory-bound	High memory bandwidth	TransData ops	Reduce transfer
Compute-bound	High AI_CORE util	Large matmul	Optimize ops
Comm-bound	HCCL ops significant	Inefficient coll	Optimize comm
Operator-fallback	AI_CPU execution	Missing NPU impl	AscendC ops

Bottleneck Diagnosis Output:

markdown

undefined

本技能包含一个AI驱动的瓶颈诊断系统，可分析性能数据自动识别根本原因：

特性:

自动根本原因分析：从性能分析数据中识别性能瓶颈
瓶颈分类：将瓶颈分为内存受限、计算受限、通信受限或算子回退类型
可执行建议：提供优先级排序的优化建议
模式匹配：检测已知的性能反模式并提出修复方案
影响评估：估算每项优化可带来的潜在性能提升

瓶颈分类:

类别	特征	原因	策略
内存受限	内存带宽占用高	TransData算子	减少数据传输
计算受限	AI_CORE利用率高	大型矩阵运算	优化算子
通信受限	HCCL算子占比显著	集合通信效率低	优化通信逻辑
算子回退	AI_CPU执行	缺少NPU实现	使用AscendC算子

瓶颈诊断输出:

markdown

undefined

Intelligent Bottleneck Diagnosis Report

智能瓶颈诊断报告

Overall Performance Summary

整体性能总结

Total Inference Time: 15.2 ms
Bottleneck Score: 78/100
Main Bottleneck Type: Memory-bound

总推理时间: 15.2 ms
瓶颈评分: 78/100
主要瓶颈类型: 内存受限

Identified Bottlenecks

已识别瓶颈

Rank	Operator	Type	Time	Percentage	Issue
1	TransData	AI_CPU	4.2 ms	27.6%	Frequent CPU-NPU transfer
2	IndexSelect	AI_CPU	2.8 ms	18.4%	Operator fallback to CPU
3	NMS	AI_CPU	1.5 ms	9.9%	No NPU implementation

排名	算子	类型	耗时	占比	问题
1	TransData	AI_CPU	4.2 ms	27.6%	频繁CPU-NPU数据传输
2	IndexSelect	AI_CPU	2.8 ms	18.4%	算子回退至CPU执行
3	NMS	AI_CPU	1.5 ms	9.9%	无NPU实现版本

Optimization Recommendations

优化建议

Priority	Operator	Issue	Solution	Expected Gain
P0	TransData	Data transfer	Reduce redundant movement	20-25%
P1	IndexSelect	CPU fallback	Implement AscendC version	15-20%
P2	NMS	CPU fallback	Use NPU-optimized NMS	10-15%

优先级	算子	问题	解决方案	预期收益
P0	TransData	数据传输	减少冗余数据移动	20-25%
P1	IndexSelect	CPU回退	实现AscendC版本	15-20%
P2	NMS	CPU回退	使用NPU优化版NMS	10-15%

Quick Wins

快速优化方案

Batch pre-processing on NPU instead of CPU
Use async data transfer with overlap
Enable memory pooling for intermediate tensors

undefined

在NPU上批量预处理而非CPU
使用异步数据传输并重叠执行
为中间张量启用内存池

undefined

Reference Documents

参考文档

Document	Description
Profiler DB Data Format	Table structure
Acceptance Criteria	Acceptance criteria
Verification Method	Verification approach
Troubleshooting	Common issues

文档	描述
Profiler DB Data Format	表结构说明
Acceptance Criteria	验收标准
Verification Method	验证方法
Troubleshooting	常见问题排查

Prerequisites

前置条件

msprof >= 7.0.0 installed
sqlite3 >= 3.0.0 installed
Have Ascend PyTorch Profiler or msprof generated database file

已安装msprof >= 7.0.0
已安装sqlite3 >= 3.0.0
拥有Ascend PyTorch Profiler或msprof生成的数据库文件

Core Commands

核心命令

bash

undefined

bash

undefined

Query operator time consumption

查询算子耗时

python3 scripts/query_profiler_db.py
--db /path/to/ascend_pytorch_profiler.db
--query "Top 10 operators by time consumption"

undefined

python3 scripts/query_profiler_db.py
--db /path/to/ascend_pytorch_profiler.db
--query "Top 10 operators by time consumption"

undefined

Parameter Confirmation

参数确认

Parameter	Description	Required
db	Profiler database path	Yes
query	Natural language query	Yes
output	Output format	No

参数	描述	是否必填
db	性能分析数据库路径	是
query	自然语言查询语句	是
output	输出格式	否