Persistence Yesterday

• Databases with a few million records on huge, single mainframe computers
• Shared database integration
• No real-time querying necessary, overnight analysis
• Develop once, use for years
• ACID
• Data-first
• Impedance mismatch

The Needs Today

• Billions of records for millions of users - mass data storage
• Impedance match
• Cheaper hardware; scale horizontally vs vertically
• Distributed across geographical regions
• Fail-over systems, clusters; sharding, master-slave replication
• Evolving architecture, software; faster development times, developer productivity, schemaless design

the Needs Today

• More specific use-cases
• Real-time queries in thousands of requests each second
• Highly connected data
• More data-models, aggregates
• ACID, CAP, Eventual Consistency
• Detail analytics, map-reduce
• Data access-firs, aggregates

EXAMPLE: E-COMMERCE PLATFORM

• Inventory
• Orders
• Shopping cart
• Session data
• Search and recommendations
• Social add-ons
• Analytics

all relational

Polyglot

Document DB Data Model

Document DB Data Model Alternative

            
{
    "customer": {
        "id": 1,
        "name": "Martin",
        "billingAddress": [{ "city": "Chicago" }],
        "orders": [{
            "id": 99,
            "customerId": 1,
            "orderItems": [{
                "productId": 27,
                "price": 32.45,
                "productName": "NoSQL Distilled"
            }],
            "shippingAddress": [{ "city": "Chicago" }],
            "orderPayment": [{
                "ccinfo": "1000-1000-1000-1000",
                "txnId": "abelif879rft",
                "billingAddress": { "city": "Chicago" }
            }]
        }]
    }
}
        

 

graph db model

MongoDB

• Productivity boost, JSON based
• Scalable on cheap hardware
• SQL-like querying capabilities
• Document-based data storage
• Schemaless design
• Aggregates and low-connected data

MongoDB Queries

SQL

MongoDB

Redis

• Fast. Very fast.
• Small and powerful data structures,
  lists, sets, hashes
• Assembly like instructions
• REST API
• In-memory
• Pub/sub
• Item expiration

Redis Instructions

http://try.redis.io             http://redis.io/topics/twitter-clone

• SET key value
• GET key
• DEL key
• INCR key
• EXPIRE key
• RPUSH (right push) key value
• LPUSH (left push) key value
• LRANGE key start end

• SADD (set add) key value
• SREM key
• SMEMBERS key
• ZADD (sorted sets) key score value
• HSET (hash set) key value
• HMSET (multi set) key field value …
• HINCRBY key field

Neo4J

• Data is mostly a set of connections, and it's those connections what matter most
• Properties on connections
• Very fast on graph modelling and search
• REST API

Example Graph

Neo4J queries