""" 详细性能分析 - 确认每个环节的耗时 """ import asyncio import sys import time from pathlib import Path sys.path.insert(0, str(Path(__file__).parent.parent)) from sqlalchemy.ext.asyncio import AsyncSession, create_async_engine from sqlalchemy.orm import sessionmaker from app.services.mid.kb_search_dynamic_tool import KbSearchDynamicTool, KbSearchDynamicConfig from app.core.config import get_settings from app.core.qdrant_client import get_qdrant_client from app.services.embedding import get_embedding_provider async def profile_detailed(): """详细分析每个环节的耗时""" settings = get_settings() print("=" * 80) print("详细性能分析") print("=" * 80) query = "三年级语文学习" tenant_id = "szmp@ash@2026" metadata_filter = {"grade": "三年级", "subject": "语文"} # 1. Embedding 生成（应该已预初始化） print("\n📊 1. Embedding 生成") print("-" * 80) start = time.time() embedding_service = await get_embedding_provider() init_time = (time.time() - start) * 1000 start = time.time() embedding_result = await embedding_service.embed_query(query) embed_time = (time.time() - start) * 1000 # 获取 embedding 向量 if hasattr(embedding_result, 'embedding_full'): query_vector = embedding_result.embedding_full elif hasattr(embedding_result, 'embedding'): query_vector = embedding_result.embedding else: query_vector = embedding_result print(f" 获取服务实例: {init_time:.2f} ms") print(f" Embedding 生成: {embed_time:.2f} ms") print(f" 向量维度: {len(query_vector)}") # 2. 获取 collections 列表（带缓存） print("\n📊 2. 获取 collections 列表") print("-" * 80) client = await get_qdrant_client() qdrant_client = await client.get_client() start = time.time() from app.services.metadata_cache_service import get_metadata_cache_service cache_service = await get_metadata_cache_service() cache_key = f"collections:{tenant_id}" # 尝试从缓存获取 redis_client = await cache_service._get_redis() cache_hit = False if redis_client and cache_service._enabled: cached = await redis_client.get(cache_key) if cached: import json tenant_collections = json.loads(cached) cache_hit = True cache_time = (time.time() - start) * 1000 print(f" ✅ 缓存命中: {cache_time:.2f} ms") print(f" Collections: {tenant_collections}") if not cache_hit: import json # 从 Qdrant 查询 start = time.time() collections = await qdrant_client.get_collections() safe_tenant_id = tenant_id.replace('@', '_') prefix = f"kb_{safe_tenant_id}" tenant_collections = [ c.name for c in collections.collections if c.name.startswith(prefix) ] tenant_collections.sort() db_time = (time.time() - start) * 1000 print(f" ❌ 缓存未命中，从 Qdrant 查询: {db_time:.2f} ms") print(f" Collections: {tenant_collections}") # 缓存结果 if redis_client and cache_service._enabled: await redis_client.setex(cache_key, 300, json.dumps(tenant_collections)) print(f" 已缓存到 Redis") # 3. Qdrant 搜索（每个 collection） print("\n📊 3. Qdrant 搜索") print("-" * 80) from qdrant_client.models import FieldCondition, Filter, MatchValue # 构建 filter start = time.time() must_conditions = [] for key, value in metadata_filter.items(): field_path = f"metadata.{key}" condition = FieldCondition( key=field_path, match=MatchValue(value=value), ) must_conditions.append(condition) qdrant_filter = Filter(must=must_conditions) if must_conditions else None filter_time = (time.time() - start) * 1000 print(f" 构建 filter: {filter_time:.2f} ms") # 逐个 collection 搜索 total_search_time = 0 for collection_name in tenant_collections: print(f"\n Collection: {collection_name}") # 检查是否存在 start = time.time() exists = await qdrant_client.collection_exists(collection_name) check_time = (time.time() - start) * 1000 print(f" 检查存在: {check_time:.2f} ms") if not exists: print(f" ❌ 不存在") continue # 搜索 start = time.time() try: results = await qdrant_client.query_points( collection_name=collection_name, query=query_vector, using="full", limit=5, score_threshold=0.5, query_filter=qdrant_filter, ) search_time = (time.time() - start) * 1000 total_search_time += search_time print(f" 搜索时间: {search_time:.2f} ms") print(f" 结果数: {len(results.points)}") except Exception as e: print(f" ❌ 搜索失败: {e}") print(f"\n 总搜索时间: {total_search_time:.2f} ms") # 4. 完整 KB Search 流程 print("\n📊 4. 完整 KB Search 流程") print("-" * 80) engine = create_async_engine(settings.database_url) async_session = sessionmaker(engine, class_=AsyncSession, expire_on_commit=False) async with async_session() as session: config = KbSearchDynamicConfig( enabled=True, top_k=5, timeout_ms=30000, min_score_threshold=0.5, ) tool = KbSearchDynamicTool(session=session, config=config) start = time.time() result = await tool.execute( query=query, tenant_id=tenant_id, scene="学习方案", top_k=5, context=metadata_filter, ) total_time = (time.time() - start) * 1000 print(f" 总耗时: {total_time:.2f} ms") print(f" 工具内部耗时: {result.duration_ms} ms") print(f" 结果数: {len(result.hits)}") # 5. 总结 print("\n" + "=" * 80) print("📈 耗时总结") print("=" * 80) print(f"\n各环节耗时:") print(f" Embedding 获取服务: {init_time:.2f} ms") print(f" Embedding 生成: {embed_time:.2f} ms") print(f" Collections 获取: {cache_time if cache_hit else db_time:.2f} ms") print(f" Filter 构建: {filter_time:.2f} ms") print(f" Qdrant 搜索: {total_search_time:.2f} ms") print(f" 完整流程: {total_time:.2f} ms") other_time = total_time - embed_time - (cache_time if cache_hit else db_time) - filter_time - total_search_time print(f" 其他开销: {other_time:.2f} ms") print("\n" + "=" * 80) if __name__ == "__main__": asyncio.run(profile_detailed())