243 lines
9.5 KiB
Python
243 lines
9.5 KiB
Python
|
# -*- coding: utf-8 -*-
|
|||
|
|
"""
|
|||
|
|
性能监控工具
|
|||
|
|
监控系统性能，识别瓶颈，提供优化建议
|
|||
|
|
"""
|
|||
|
|
|
|||
|
|
import time
|
|||
|
|
import logging
|
|||
|
|
import threading
|
|||
|
|
from typing import Dict, List, Any, Optional
|
|||
|
|
from datetime import datetime, timedelta
|
|||
|
|
from collections import defaultdict, deque
|
|||
|
|
import psutil
|
|||
|
|
|
|||
|
|
from .performance_config import PerformanceConfig
|
|||
|
|
|
|||
|
|
logger = logging.getLogger(__name__)
|
|||
|
|
|
|||
|
|
class PerformanceMonitor:
|
|||
|
|
"""性能监控器"""
|
|||
|
|
|
|||
|
|
def __init__(self):
|
|||
|
|
self.query_times = deque(maxlen=1000)
|
|||
|
|
self.api_response_times = deque(maxlen=1000)
|
|||
|
|
self.cache_hit_rates = defaultdict(list)
|
|||
|
|
self.system_metrics = deque(maxlen=100)
|
|||
|
|
|
|||
|
|
self.monitoring_enabled = True
|
|||
|
|
self.monitor_thread = None
|
|||
|
|
self.start_monitoring()
|
|||
|
|
|
|||
|
|
def start_monitoring(self):
|
|||
|
|
"""启动性能监控"""
|
|||
|
|
if self.monitor_thread and self.monitor_thread.is_alive():
|
|||
|
|
return
|
|||
|
|
|
|||
|
|
self.monitoring_enabled = True
|
|||
|
|
self.monitor_thread = threading.Thread(target=self._monitor_loop, daemon=True)
|
|||
|
|
self.monitor_thread.start()
|
|||
|
|
logger.info("性能监控已启动")
|
|||
|
|
|
|||
|
|
def stop_monitoring(self):
|
|||
|
|
"""停止性能监控"""
|
|||
|
|
self.monitoring_enabled = False
|
|||
|
|
if self.monitor_thread:
|
|||
|
|
self.monitor_thread.join(timeout=5)
|
|||
|
|
logger.info("性能监控已停止")
|
|||
|
|
|
|||
|
|
def _monitor_loop(self):
|
|||
|
|
"""监控循环"""
|
|||
|
|
while self.monitoring_enabled:
|
|||
|
|
try:
|
|||
|
|
self._collect_system_metrics()
|
|||
|
|
self._analyze_performance()
|
|||
|
|
time.sleep(PerformanceConfig.MONITORING_INTERVAL)
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"性能监控异常: {e}")
|
|||
|
|
time.sleep(10)
|
|||
|
|
|
|||
|
|
def _collect_system_metrics(self):
|
|||
|
|
"""收集系统指标"""
|
|||
|
|
try:
|
|||
|
|
metrics = {
|
|||
|
|
'timestamp': datetime.now(),
|
|||
|
|
'cpu_percent': psutil.cpu_percent(interval=1),
|
|||
|
|
'memory_percent': psutil.virtual_memory().percent,
|
|||
|
|
'disk_percent': psutil.disk_usage('/').percent,
|
|||
|
|
'active_connections': len(psutil.net_connections()),
|
|||
|
|
'query_count': len(self.query_times),
|
|||
|
|
'api_count': len(self.api_response_times)
|
|||
|
|
}
|
|||
|
|
self.system_metrics.append(metrics)
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"收集系统指标失败: {e}")
|
|||
|
|
|
|||
|
|
def _analyze_performance(self):
|
|||
|
|
"""分析性能"""
|
|||
|
|
try:
|
|||
|
|
# 分析查询性能
|
|||
|
|
if self.query_times:
|
|||
|
|
avg_query_time = sum(self.query_times) / len(self.query_times)
|
|||
|
|
slow_queries = [t for t in self.query_times if t > PerformanceConfig.SLOW_QUERY_THRESHOLD]
|
|||
|
|
|
|||
|
|
if len(slow_queries) > len(self.query_times) * 0.1: # 超过10%的查询是慢查询
|
|||
|
|
logger.warning(f"检测到慢查询: 平均{avg_query_time:.2f}s, 慢查询比例{len(slow_queries)/len(self.query_times)*100:.1f}%")
|
|||
|
|
|
|||
|
|
# 分析API性能
|
|||
|
|
if self.api_response_times:
|
|||
|
|
avg_api_time = sum(self.api_response_times) / len(self.api_response_times)
|
|||
|
|
if avg_api_time > 2.0: # 平均API响应时间超过2秒
|
|||
|
|
logger.warning(f"API响应时间较慢: 平均{avg_api_time:.2f}s")
|
|||
|
|
|
|||
|
|
# 分析缓存性能
|
|||
|
|
for cache_name, hit_rates in self.cache_hit_rates.items():
|
|||
|
|
if hit_rates:
|
|||
|
|
avg_hit_rate = sum(hit_rates) / len(hit_rates)
|
|||
|
|
if avg_hit_rate < 0.5: # 缓存命中率低于50%
|
|||
|
|
logger.warning(f"缓存命中率较低: {cache_name} {avg_hit_rate*100:.1f}%")
|
|||
|
|
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"性能分析失败: {e}")
|
|||
|
|
|
|||
|
|
def record_query_time(self, query_name: str, duration: float):
|
|||
|
|
"""记录查询时间"""
|
|||
|
|
self.query_times.append(duration)
|
|||
|
|
|
|||
|
|
if PerformanceConfig.PERFORMANCE_LOG_ENABLED:
|
|||
|
|
if duration > PerformanceConfig.SLOW_QUERY_THRESHOLD:
|
|||
|
|
logger.warning(f"慢查询: {query_name} 耗时 {duration:.2f}s")
|
|||
|
|
else:
|
|||
|
|
logger.debug(f"查询: {query_name} 耗时 {duration:.2f}s")
|
|||
|
|
|
|||
|
|
def record_api_response_time(self, api_name: str, duration: float):
|
|||
|
|
"""记录API响应时间"""
|
|||
|
|
self.api_response_times.append(duration)
|
|||
|
|
|
|||
|
|
if PerformanceConfig.PERFORMANCE_LOG_ENABLED:
|
|||
|
|
if duration > 2.0:
|
|||
|
|
logger.warning(f"慢API: {api_name} 耗时 {duration:.2f}s")
|
|||
|
|
else:
|
|||
|
|
logger.debug(f"API: {api_name} 耗时 {duration:.2f}s")
|
|||
|
|
|
|||
|
|
def record_cache_hit(self, cache_name: str, hit: bool):
|
|||
|
|
"""记录缓存命中"""
|
|||
|
|
hit_rate = 1.0 if hit else 0.0
|
|||
|
|
self.cache_hit_rates[cache_name].append(hit_rate)
|
|||
|
|
|
|||
|
|
# 保持最近100次记录
|
|||
|
|
if len(self.cache_hit_rates[cache_name]) > 100:
|
|||
|
|
self.cache_hit_rates[cache_name] = self.cache_hit_rates[cache_name][-100:]
|
|||
|
|
|
|||
|
|
def get_performance_report(self) -> Dict[str, Any]:
|
|||
|
|
"""获取性能报告"""
|
|||
|
|
try:
|
|||
|
|
report = {
|
|||
|
|
'timestamp': datetime.now().isoformat(),
|
|||
|
|
'query_performance': self._get_query_performance(),
|
|||
|
|
'api_performance': self._get_api_performance(),
|
|||
|
|
'cache_performance': self._get_cache_performance(),
|
|||
|
|
'system_performance': self._get_system_performance(),
|
|||
|
|
'recommendations': self._get_optimization_recommendations()
|
|||
|
|
}
|
|||
|
|
return report
|
|||
|
|
except Exception as e:
|
|||
|
|
logger.error(f"生成性能报告失败: {e}")
|
|||
|
|
return {'error': str(e)}
|
|||
|
|
|
|||
|
|
def _get_query_performance(self) -> Dict[str, Any]:
|
|||
|
|
"""获取查询性能"""
|
|||
|
|
if not self.query_times:
|
|||
|
|
return {'status': 'no_data'}
|
|||
|
|
|
|||
|
|
avg_time = sum(self.query_times) / len(self.query_times)
|
|||
|
|
max_time = max(self.query_times)
|
|||
|
|
slow_queries = len([t for t in self.query_times if t > PerformanceConfig.SLOW_QUERY_THRESHOLD])
|
|||
|
|
|
|||
|
|
return {
|
|||
|
|
'total_queries': len(self.query_times),
|
|||
|
|
'avg_time': round(avg_time, 3),
|
|||
|
|
'max_time': round(max_time, 3),
|
|||
|
|
'slow_queries': slow_queries,
|
|||
|
|
'slow_query_rate': round(slow_queries / len(self.query_times) * 100, 1)
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
def _get_api_performance(self) -> Dict[str, Any]:
|
|||
|
|
"""获取API性能"""
|
|||
|
|
if not self.api_response_times:
|
|||
|
|
return {'status': 'no_data'}
|
|||
|
|
|
|||
|
|
avg_time = sum(self.api_response_times) / len(self.api_response_times)
|
|||
|
|
max_time = max(self.api_response_times)
|
|||
|
|
slow_apis = len([t for t in self.api_response_times if t > 2.0])
|
|||
|
|
|
|||
|
|
return {
|
|||
|
|
'total_requests': len(self.api_response_times),
|
|||
|
|
'avg_time': round(avg_time, 3),
|
|||
|
|
'max_time': round(max_time, 3),
|
|||
|
|
'slow_requests': slow_apis,
|
|||
|
|
'slow_request_rate': round(slow_apis / len(self.api_response_times) * 100, 1)
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
def _get_cache_performance(self) -> Dict[str, Any]:
|
|||
|
|
"""获取缓存性能"""
|
|||
|
|
cache_stats = {}
|
|||
|
|
for cache_name, hit_rates in self.cache_hit_rates.items():
|
|||
|
|
if hit_rates:
|
|||
|
|
avg_hit_rate = sum(hit_rates) / len(hit_rates)
|
|||
|
|
cache_stats[cache_name] = {
|
|||
|
|
'hit_rate': round(avg_hit_rate * 100, 1),
|
|||
|
|
'total_requests': len(hit_rates)
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
return cache_stats if cache_stats else {'status': 'no_data'}
|
|||
|
|
|
|||
|
|
def _get_system_performance(self) -> Dict[str, Any]:
|
|||
|
|
"""获取系统性能"""
|
|||
|
|
if not self.system_metrics:
|
|||
|
|
return {'status': 'no_data'}
|
|||
|
|
|
|||
|
|
latest = self.system_metrics[-1]
|
|||
|
|
return {
|
|||
|
|
'cpu_percent': latest['cpu_percent'],
|
|||
|
|
'memory_percent': latest['memory_percent'],
|
|||
|
|
'disk_percent': latest['disk_percent'],
|
|||
|
|
'active_connections': latest['active_connections']
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
def _get_optimization_recommendations(self) -> List[str]:
|
|||
|
|
"""获取优化建议"""
|
|||
|
|
recommendations = []
|
|||
|
|
|
|||
|
|
# 查询性能建议
|
|||
|
|
if self.query_times:
|
|||
|
|
avg_query_time = sum(self.query_times) / len(self.query_times)
|
|||
|
|
if avg_query_time > 1.0:
|
|||
|
|
recommendations.append("考虑优化数据库查询，添加索引或使用缓存")
|
|||
|
|
|
|||
|
|
# API性能建议
|
|||
|
|
if self.api_response_times:
|
|||
|
|
avg_api_time = sum(self.api_response_times) / len(self.api_response_times)
|
|||
|
|
if avg_api_time > 2.0:
|
|||
|
|
recommendations.append("考虑优化API响应时间，使用异步处理或缓存")
|
|||
|
|
|
|||
|
|
# 缓存性能建议
|
|||
|
|
for cache_name, hit_rates in self.cache_hit_rates.items():
|
|||
|
|
if hit_rates:
|
|||
|
|
avg_hit_rate = sum(hit_rates) / len(hit_rates)
|
|||
|
|
if avg_hit_rate < 0.5:
|
|||
|
|
recommendations.append(f"优化{cache_name}缓存策略，提高命中率")
|
|||
|
|
|
|||
|
|
# 系统资源建议
|
|||
|
|
if self.system_metrics:
|
|||
|
|
latest = self.system_metrics[-1]
|
|||
|
|
if latest['cpu_percent'] > 80:
|
|||
|
|
recommendations.append("CPU使用率过高，考虑优化计算密集型操作")
|
|||
|
|
if latest['memory_percent'] > 80:
|
|||
|
|
recommendations.append("内存使用率过高，考虑清理缓存或优化内存使用")
|
|||
|
|
|
|||
|
|
return recommendations
|
|||
|
|
|
|||
|
|
# 全局性能监控器实例
|
|||
|
|
performance_monitor = PerformanceMonitor()
|