NoSQL MongoDB(1)

NoSQL databases currently in use can be divided into four broad categories:

1.Key-value data stores: Data is stored in key-value pairs, and values ​​are retrieved through keys. For example redis, dynomite,
voldemort.

2.Column-based databases:
These databases organize data stored in tables, similar to RDBMS. However, they store content in columns rather than rows. Very good for data warehouse applications. For example Hbase,
Cassandra, Hypertable.

3.Document-based databases:
Data is stored and organized as a collection document. These documents are flexible and can have any number of fields per document. For example CouchDB, MongoDB.

4.Graph-based data-stores:
Such databases are used for data storage and retrieval in computer science graph theory. Focus on data interconnection in different parts. The units of data are visualized as nodes and the relationships between them are defined as edges connecting the nodes. For example Neo4j.

MongoDB components:

1.Database:
There can be multiple databases, each acting as an independent container. Each database can contain one or more collections.

2.Collection:
A collection is a group of documents. Logically equivalent to a table in a relational database. But unlike tables, there is no need to define a data structure beforehand when storing data in a collection.

3.Document:
The unit in which documents are stored in the collection. A document contains a series of fields or key-value pairs. The keys are strings and the values ​​can be of various types: strings, integers, floats, timestamps, booleans, etc. It can even store another document. It is stored in json format.

MongoDB data exchange format:

The structure of the document is a JSON object. When this document is stored in the database, it will be serialized into a special binary encoding format, called BSON.

BSON is MongoDB’s default data exchange format. Compared with XML and JSON, the advantage of BSON is that it is more efficient in memory consumption and processing time. JSON supports all data types supported by BSON, and also supports some special data types, such as regular expressions, object IDs, dates, binary data and codes. Better portability, programming languages ​​such as PHP, PYTHON, JAVA, etc. can be directly converted from BSON format, making it easy to integrate with MongoDB and manage and maintain data.

Comparison between MongoDB and RDMS:

RDMS: database—>table—>row—>column

MongoDB: database—>collection—>document—>field

There are no foreign keys in MongoDB collections, no join queries, and constraint management is usually handled by the application layer. Therefore, its architecture is very flexible and there is no expensive alter.
table operations.

MongoDB installation:

# apt-key adv ?keyserver keyserver.ubuntu.com ?recv 7F0CEB10

# vi /etc/apt/sources.list.d/10gen.list

deb http://downloads-distro.mongodb.org/repo/debian-sysvinit
dist 10gen

# apt-get update

# apt-get install mongodb-10gen

Command line operations:

#mongo

> use myfirstdb

> db.movies.insert({name:”Hangover”, genre:”comedy”,
year:2010})

> db.movies.find()

{ “_id” : ObjectId(“4fe1680418f0c60b236a4f28″), “name” :
“Hangover”, “genre” : “comedy”, “year” : 2010 }

PHP-MongoDB extension:

# apt-get install php-pear

# pecl install mongo

# vim /etc/php5/conf.d/mongo.ini

extension=mongo.so

Use PHP to operate MongoDB:

#vimongodb.php

<?php

try{

$mOngo= new
Mongo(“mongodb://127.0.0.1:27017″,array(‘timeout’=>100));
//Connect to mongodb

$databases = $mongo->listDBs(); //List all databases

print_r($databases);

$mongo->close();

}catch (Exception $e){

die ($e->getMessage());

}

?>

The query results are as follows:

Array

(

[databases] => Array

(

[0] => Array

(

[name] => myfirstdb

[sizeOnDisk] => 218103808

[empty] =>

)

[1] => Array

(

[name] => local

[sizeOnDisk] => 1

[empty] => 1

)

[2] => Array

(

[name] => test

[sizeOnDisk] => 1

[empty] => 1

)

)

[totalSize] => 218103808

[ok] => 1

)