aiemu/ddbs

A naive distributed database management system.

DDBS: A manual

Introduction

This is a distributed database system based on mongodb, redis and hadoop file system. At the same time, vue and flask are used to realize the front and back ends for the convenience of users. In terms of monitoring, MongoDB Compass and Hadoop Web Page are used to monitor mongodb and hadoop respectively.

Quick Start

Prerequisites

• Docker 20.10.16
• MongoDB shell version v5.0.9
• redis-cli 6.2.6
• python 3.8
• node.js 16.9.0
• npm 7.22.0

One-step Start

You can quickly start by running the command below, or follow the prompts to start step by step (Note: Please modify the privateIP and PRIVATE_IP in the project to your LAN ip before execution).

Backend

Execute the following commands in src/ddbs-construct/

chmod +x ddbs.sh
./ddbs.sh

Frontend

Execute the following commands in src/ddbs-manage/

chmod +x ddbs-manage.sh
./ddbs-manage.sh

Python Package

The pip package required to run the code is installed by executing the following command in the src:

python -m pip install -r requirements.txt

Docker Setup

MongoDB

First we use docker to start the mongodb cluster, including four shard servers, two config servers and one router server. Execute the following command to start the cluster (Note: Please modify the privateIP in docker/mongo-docker-compose.yml to your LAN ip before execution):

docker-compose -f docker/mongo-docker-compose.yml up -d

Next, we configure each server separately. First we execute the following command to configure the config server (Note: Please modify the privateIP in configsvr.js to your LAN ip before execution):

mongo mongodb://localhost:27600 docker/scripts/configsvr.js

After configuring the config server, we execute the following commands to configure four shard servers (Note: Please modify the privateIP in shardrep.js to your LAN ip before execution):

mongo mongodb://localhost:27700 docker/scripts/shard0.js
mongo mongodb://localhost:27711 docker/scripts/shard1.js
mongo mongodb://localhost:27701 docker/scripts/shardrep.js

Finally, let's configure the router server of this mongodb cluster:

mongo mongodb://localhost:27500 docker/scripts/router.js

Through the above steps, we have completed the construction of the mongodb cluster. This cluster includes two databases, DBMS1 and DBMS2, and some backups. The next step is to store the data in the cluster

python init_db.py

Redis

First we start a docker containing redis using the following command

docker-compose -f docker/redis-docker-compose.yml up -d

Then use the following command to modify the redis configuration (Note: Please modify the PRIVATE_IP to your LAN ip before execution):

cat docker/scripts/redis.txt | redis-cli -h <PRIVATE_IP> -p 6379

HDFS

Before building the docker image of hdfs, you should first put all the generated articles in the directory hadoop/articles/. Then execute the following command to build the docker image of hdfs, which includes two slaves and one master:

docker build -t hadoop:3.2.1 .

Building hadoop can be a lengthy process that can take hours. After the build is complete, execute the following command to create a hadoop cluster

chmod +x docker/scripts/init_hdfs.sh
./docker/scripts/init_hdfs.sh

Start Backend

Next we start the flask-based backend application with following command:

python app.py runserver

Then, the backend will be started at http://127.0.0.1:5001.

Start Frontend

The front end is done with Vue, first we need to install the dependency library and then start the frontend.

cd ../ddbs-manage
npm install
npm run dev

Now, you can visit the frontend at http://127.0.0.1:8080 like this:

Monitor

We use mongodb compass and hadoop web page to monitor the status of mongodb and hadoop.

Mongodb Compass

Hadoop Web page