MapReduce在百度百科中的解释是：

MapReduce是一种编程模型，用于大规模数据集（大于1TB）的并行运算。"Map（映射）"和"Reduce（归约）"是它们的主要思想，都是从函数式编程语言和矢量编程语言借来的特性。它极大地方便了编程人员在不会分布式并行编程的情况下，将自己的程序运行在分布式系统上。当前的软件实现是指定一个Map（映射）函数，用来把一组键值对映射成一组新的键值对，指定并发的Reduce（归约）函数，用来保证所有映射的键值对中的每一个共享相同的键组。

通过这段描述，我们可以明确知道MapReduce是面向大数据并行处理的计算模型、框架和平台，在传统学习中，通常会在Hadoop等分布式框架下进行MapReduce相关工作，随着云计算的逐渐发展，各个云厂商也都先后推出了在线的MapReduce业务。

本文我们将通过MapReduce模型实现一个简单的WordCount算法，区别于传统使用Hadoop等大数据框架，我们使用的是对象存储与云函数的结合。

理论基础

在开始之前，我们根据MapReduce要求先绘制一个简单的流程图:

在这个结构中，我们需要2个云函数分别作Mapper和Reducer，3个对象存储的存储桶分别作为输入的存储桶、中间临时缓存的存储桶以及结果存储桶。在开始实践前，我们先在广州区准备3个对象存储：

对象存储1	ap-guangzhou	srcmr
对象存储2	ap-guangzhou	middlestagebucket
对象存储3	ap-guangzhou	destcmr

为了让整个Mapper和Reducer逻辑更加清晰，我们先对传统的WordCount结构进行改造，使其更加适合云函数，同时合理分配Mapper和Reducer的工作：

功能实现

编写Mapper相关逻辑：

# -*- coding: utf8 -*-
import datetime
from qcloud_cos_v5 import CosConfig
from qcloud_cos_v5 import CosS3Client
from qcloud_cos_v5 import CosServiceError
import re
import os
import sys
import logging
logging.basicConfig(level=logging.INFO, stream=sys.stdout)
logger = logging.getLogger()
logger.setLevel(level=logging.INFO)
region = u'ap-guangzhou'  # 根据实际情况，修改地域
middle_stage_bucket = 'middlestagebucket'  # 根据实际情况，修改bucket名
def delete_file_folder(src):
    if os.path.isfile(src):
        try:
            os.remove(src)
        except:
            pass
    elif os.path.isdir(src):
        for item in os.listdir(src):
            itemsrc = os.path.join(src, item)
            delete_file_folder(itemsrc)
        try:
            os.rmdir(src)
        except:
            pass
def download_file(cos_client, bucket, key, download_path):
    logger.info("Get from [%s] to download file [%s]" % (bucket, key))
    try:
        response = cos_client.get_object(Bucket=bucket, Key=key, )
        response['Body'].get_stream_to_file(download_path)
    except CosServiceError as e:
        print(e.get_error_code())
        print(e.get_error_msg())
        return -1
    return 0
def upload_file(cos_client, bucket, key, local_file_path):
    logger.info("Start to upload file to cos")
    try:
        response = cos_client.put_object_from_local_file(
            Bucket=bucket,
            LocalFilePath=local_file_path,
            Key='{}'.format(key))
    except CosServiceError as e:
        print(e.get_error_code())
        print(e.get_error_msg())
        return -1
    logger.info("Upload data map file [%s] Success" % key)
    return 0
def do_mapping(cos_client, bucket, key, middle_stage_bucket, middle_file_key):
    src_file_path = u'/tmp/' + key.split('/')[-1]
    middle_file_path = u'/tmp/' + u'mapped_' + key.split('/')[-1]
    download_ret = download_file(cos_client, bucket, key, src_file_path)  # download src file
    if download_ret == 0:
        inputfile = open(src_file_path, 'r')  # open local /tmp file
        mapfile = open(middle_file_path, 'w')  # open a new file write stream
        for line in inputfile:
            line = re.sub('[^a-zA-Z0-9]', ' ', line)  # replace non-alphabetic/number characters
            words = line.split()
            for word in words:
                mapfile.write('%s\t%s' % (word, 1))  # count for 1
                mapfile.write('\n')
        inputfile.close()
        mapfile.close()
        upload_ret = upload_file(cos_client, middle_stage_bucket, middle_file_key,
                                 middle_file_path)  # upload the file's each word
        delete_file_folder(src_file_path)
        delete_file_folder(middle_file_path)
        return upload_ret
    else:
        return -1
def map_caller(event, context, cos_client):
    appid = event['Records'][0]['cos']['cosBucket']['appid']
    bucket = event['Records'][0]['cos']['cosBucket']['name'] + '-' + appid
    key = event['Records'][0]['cos']['cosObject']['key']
    key = key.replace('/' + str(appid) + '/' + event['Records'][0]['cos']['cosBucket']['name'] + '/', '', 1)
    logger.info("Key is " + key)
    middle_bucket = middle_stage_bucket + '-' + appid
    middle_file_key = '/' + 'middle_' + key.split('/')[-1]
    return do_mapping(cos_client, bucket, key, middle_bucket, middle_file_key)
def main_handler(event, context):
    logger.info("start main handler")
    if "Records" not in event.keys():
        return {"errorMsg": "event is not come from cos"}
    secret_id = "" 
    secret_key = ""  
    config = CosConfig(Region=region, SecretId=secret_id, SecretKey=secret_key, )
    cos_client = CosS3Client(config)
    start_time = datetime.datetime.now()
    res = map_caller(event, context, cos_client)
    end_time = datetime.datetime.now()
    print("data mapping duration: " + str((end_time - start_time).microseconds / 1000) + "ms")
    if res == 0:
        return "Data mapping SUCCESS"
    else:
        return "Data mapping FAILED"

同样的方法，建立reducer.py文件，编写Reducer逻辑：

# -*- coding: utf8 -*-
from qcloud_cos_v5 import CosConfig
from qcloud_cos_v5 import CosS3Client
from qcloud_cos_v5 import CosServiceError
from operator import itemgetter
import os
import sys
import datetime
import logging
region = u'ap-guangzhou'  # 根据实际情况，修改地域
result_bucket = u'destmr'  # 根据实际情况，修改bucket名
logging.basicConfig(level=logging.INFO, stream=sys.stdout)
logger = logging.getLogger()
logger.setLevel(level=logging.INFO)
def delete_file_folder(src):
    if os.path.isfile(src):
        try:
            os.remove(src)
        except:
            pass
    elif os.path.isdir(src):
        for item in os.listdir(src):
            itemsrc = os.path.join(src, item)
            delete_file_folder(itemsrc)
        try:
            os.rmdir(src)
        except:
            pass
def download_file(cos_client, bucket, key, download_path):
    logger.info("Get from [%s] to download file [%s]" % (bucket, key))
    try:
        response = cos_client.get_object(Bucket=bucket, Key=key, )
        response['Body'].get_stream_to_file(download_path)
    except CosServiceError as e:
        print(e.get_error_code())
        print(e.get_error_msg())
        return -1
    return 0
def upload_file(cos_client, bucket, key, local_file_path):
    logger.info("Start to upload file to cos")
    try:
        response = cos_client.put_object_from_local_file(
            Bucket=bucket,
            LocalFilePath=local_file_path,
            Key='{}'.format(key))
    except CosServiceError as e:
        print(e.get_error_code())
        print(e.get_error_msg())
        return -1
    logger.info("Upload data map file [%s] Success" % key)
    return 0
def qcloud_reducer(cos_client, bucket, key, result_bucket, result_key):
    word2count = {}
    src_file_path = u'/tmp/' + key.split('/')[-1]
    result_file_path = u'/tmp/' + u'result_' + key.split('/')[-1]
    download_ret = download_file(cos_client, bucket, key, src_file_path)
    if download_ret == 0:
        map_file = open(src_file_path, 'r')
        result_file = open(result_file_path, 'w')
        for line in map_file:
            line = line.strip()
            word, count = line.split('\t', 1)
            try:
                count = int(count)
                word2count[word] = word2count.get(word, 0) + count
            except ValueError:
                logger.error("error value: %s, current line: %s" % (ValueError, line))
                continue
        map_file.close()
        delete_file_folder(src_file_path)
    sorted_word2count = sorted(word2count.items(), key=itemgetter(1))[::-1]
    for wordcount in sorted_word2count:
        res = '%s\t%s' % (wordcount[0], wordcount[1])
        result_file.write(res)
        result_file.write('\n')
    result_file.close()
    upload_ret = upload_file(cos_client, result_bucket, result_key, result_file_path)
    delete_file_folder(result_file_path)
    return upload_ret
def reduce_caller(event, context, cos_client):
    appid = event['Records'][0]['cos']['cosBucket']['appid']
    bucket = event['Records'][0]['cos']['cosBucket']['name'] + '-' + appid
    key = event['Records'][0]['cos']['cosObject']['key']
    key = key.replace('/' + str(appid) + '/' + event['Records'][0]['cos']['cosBucket']['name'] + '/', '', 1)
    logger.info("Key is " + key)
    res_bucket = result_bucket + '-' + appid
    result_key = '/' + 'result_' + key.split('/')[-1]
    return qcloud_reducer(cos_client, bucket, key, res_bucket, result_key)
def main_handler(event, context):
    logger.info("start main handler")
    if "Records" not in event.keys():
        return {"errorMsg": "event is not come from cos"}
    secret_id = "SecretId" 
    secret_key = "SecretKey"  
    config = CosConfig(Region=region, SecretId=secret_id, SecretKey=secret_key, )
    cos_client = CosS3Client(config)
    start_time = datetime.datetime.now()
    res = reduce_caller(event, context, cos_client)
    end_time = datetime.datetime.now()
    print("data reducing duration: " + str((end_time - start_time).microseconds / 1000) + "ms")
    if res == 0:
        return "Data reducing SUCCESS"
    else:
        return "Data reducing FAILED"

部署与测试

通过Serverless Framework的yaml规范，编写serveerless.yaml:

WordCountMapper:
  component: "@serverless/tencent-scf"
  inputs:
    name: mapper
    codeUri: ./code
    handler: index.main_handler
    runtime: Python3.6
    region: ap-guangzhou
    description: 网站监控
    memorySize: 64
    timeout: 20
    events:
      - cos:
          name: srcmr-1256773370.cos.ap-guangzhou.myqcloud.com
          parameters:
            bucket: srcmr-1256773370.cos.ap-guangzhou.myqcloud.com
            filter:
              prefix: ''
              suffix: ''
            events: cos:ObjectCreated:*
            enable: true

WordCountReducer:
  component: "@serverless/tencent-scf"
  inputs:
    name: reducer
    codeUri: ./code
    handler: index.main_handler
    runtime: Python3.6
    region: ap-guangzhou
    description: 网站监控
    memorySize: 64
    timeout: 20
    events:
      - cos:
          name: middlestagebucket-1256773370.cos.ap-guangzhou.myqcloud.com
          parameters:
            bucket: middlestagebucket-1256773370.cos.ap-guangzhou.myqcloud.com
            filter:
              prefix: ''
              suffix: ''
            events: cos:ObjectCreated:*
            enable: true

完成之后，通过sls --debug指令进行部署，成功之后进行基本的测试：

首先准备一个英文文档:

登录腾讯云后台，打开最初建立的存储桶：srcmr，并上传该文件;
上传成功之后，稍等片刻就可以看到Reducer程序已经在Mapper执行之后，产出日志:

此时，打开结果存储桶，查看结果:

这样，我们就完成了简单的词频统计功能。

总结

相对来说，Serverless架构比较适合做大数据处理，在腾讯云官网对Serverless应用场景的描述就包含有数据ETL处理：

一些数据处理系统中，常常需要周期性/计划性地处理庞大的数据量。例如：证券公司每12小时统计一次该时段的交易情况并整理出该时段交易量 top 5，每天处理一遍秒杀网站的交易流日志获取因售罄而导致的错误从而分析商品热度和趋势等。云函数近乎无限扩容的能力可以使您轻松地进行大容量数据的计算。我们利用云函数可以对源数据并发执行多个 mapper 和 reducer 函数，在短时间内完成工作；相比传统的工作方式，使用云函数更能避免资源的闲置浪费从而节省资金。

通过本实例，希望读者可以对Serverless架构的应用场景有更多的启发，了解到Serverless不仅仅在监控告警方面有着很好的表现，在大数据领域也不甘落后。在实际生产中，每个项目都不会是单个函数单打独斗的，而是多个函数组合应用，形成一个Service体系，所以一键部署多个函数就显得尤为重要。

创作场景

Serverless 实战：利用函数计算与对象存储实现 WordCount