Python实现基于MapReduce思想的系统日志统计分析

Elena

发布时间：2026-10-13| 6人浏览过

原创

本文介绍使用Python实现类似MapReduce的大数据分析流程，涵盖大文件切分、Mapper映射及Reducer归约三个核心步骤。通过正则表达式提取日志日期并统计频次，结合多线程技术处理切分后的文件，最终输出按日期排序的统计结果，适用于系统日志的批量处理场景。

本文分享一个使用Python进行大数据分析及操作系统日志处理的完整示例，供读者参考学习。

一代码实现

1、大文件切分逻辑

import os
import os.path
import time
def FileSplit(sourceFile, targetFolder):
  if not os.path.isfile(sourceFile):
    print(sourceFile, ' does not exist.')
    return
  if not os.path.isdir(targetFolder):
    os.mkdir(targetFolder)
  tempData = []
  number = 1000
  fileNum = 1
  linesRead = 0
  with open(sourceFile, 'r') as srcFile:
    dataLine = srcFile.readline().strip()
    while dataLine:
      for i in range(number):
        tempData.append(dataLine)
        dataLine = srcFile.readline()
        if not dataLine:
          break
      desFile = os.path.join(targetFolder, sourceFile[0:-4] + str(fileNum) + '.txt')
      with open(desFile, 'a+') as f:
        f.writelines(tempData)
      tempData = []
      fileNum = fileNum + 1
if __name__ == '__main__':
  #sourceFile = input('Input the source file to split:')
  #targetFolder = input('Input the target folder you want to place the split files:')
  sourceFile = 'test.txt'
  targetFolder = 'test'
  FileSplit(sourceFile, targetFolder)

2、Mapper映射代码

import os
import re
import threading
import time
def Map(sourceFile):
  if not os.path.exists(sourceFile):
    print(sourceFile, ' does not exist.')
    return
  pattern = re.compile(r'[0-9]{1,2}/[0-9]{1,2}/[0-9]{4}')
  result = {}
  with open(sourceFile, 'r') as srcFile:
    for dataLine in srcFile:
      r = pattern.findall(dataLine)
      if r:
        t = result.get(r[0], 0)
        t += 1
        result[r[0]] = t
  desFile = sourceFile[0:-4] + '_map.txt'
  with open(desFile, 'a+') as fp:
    for k, v in result.items():
      fp.write(k + ':' + str(v) + '\n')
if __name__ == '__main__':
  desFolder = 'test'
  files = os.listdir(desFolder)
  #如果不使用多线程，可以直接这样写
  '''for f in files:
    Map(desFolder + '\\' + f)'''
  #使用多线程
  def Main(i):
    Map(desFolder + '\\' + files[i])
  fileNumber = len(files)
  for i in range(fileNumber):
    t = threading.Thread(target = Main, args =(i,))
    t.start()

3、Reducer归约代码

import os
def Reduce(sourceFolder, targetFile):
  if not os.path.isdir(sourceFolder):
    print(sourceFolder, ' does not exist.')
    return
  result = {}
  #Deal only with the mapped files
  allFiles = [sourceFolder+'\\'+f for f in os.listdir(sourceFolder) if f.endswith('_map.txt')]
  for f in allFiles:
    with open(f, 'r') as fp:
      for line in fp:
        line = line.strip()
        if not line:
          continue
        position = line.index(':')
        key = line[0:position]
        value = int(line[position + 1:])
        result[key] = result.get(key,0) + value
  with open(targetFile, 'w') as fp:
    for k,v in result.items():
      fp.write(k + ':' + str(v) + '\n')
if __name__ == '__main__':
  Reduce('test', 'test\\result.txt')

二运行结果展示

按顺序依次执行上述三个程序模块，可获得最终的统计结果：