NoSQL Databases

NoSQL databases (Not Only SQL) are designed for large sets of distributed data. Some big data performance issues that are not effectively handled by relational databases are easily managed by NoSQL databases. They are very efficient at analyzing large-size unstructured data that may be stored across multiple virtual servers in the cloud.

Types of NoSQL Databases

NoSQL databases are categorized into four main types based on their data model −

• Document Store − Stores data as JSON-like documents. Examples: MongoDB, CouchDB.
• Key-Value Store − Stores data as key-value pairs. Examples: Redis, DynamoDB.
• Column-Family Store − Stores data in columns grouped into families. Examples: Apache Cassandra, HBase.
• Graph Database − Stores data as nodes and edges (relationships). Examples: Neo4j, Amazon Neptune.

Characteristics

• Schema-free − No predefined schema required; data structure can vary per record.
• Horizontally scalable − Scale out by adding more servers instead of upgrading hardware.
• High performance − Optimized for read/write operations on large datasets.
• Distributed − Data is stored across multiple servers or cloud instances.
• Eventual consistency − Follows CAP theorem, prioritizing availability and partition tolerance.

NoSQL vs RDBMS

Example NoSQL Query

Here's a basic MongoDB query example to demonstrate NoSQL syntax −

-- MongoDB query to find documents
db.users.find({"age": {"$gt": 25}, "city": "New York"})

-- Insert a document
db.users.insertOne({
   "name": "John Doe",
   "age": 30,
   "email": "john@example.com",
   "skills": ["JavaScript", "Python", "NoSQL"]
})

-- Update a document
db.users.updateOne(
   {"name": "John Doe"},
   {"$set": {"age": 31}}
)

Conclusion

NoSQL databases are ideal for handling large volumes of unstructured or semi-structured data distributed across cloud servers. They offer horizontal scalability and flexible schemas, making them well-suited for big data, real-time applications, and scenarios where traditional relational databases face performance limitations.