No relational Databases

Central question

What database or combination of databases best resolves your problem?

Can I use this database to store and refine this data?

Should I?

The question you must always ask yourself

Relational (PostgreSQL)
Key-Value (Voldemort, Riak, Redis)
Columnar (HBase, Cassandra, Hypertable)
Document (MongoDB, CouchDB)
Graph (Neo4J)

Types

Key-Value

The key-value (KV) store is the simplest model we cover. KV store pairs keys to values in much the same way that a map

Minuses:

doesn't support complex query and aggregation

Pluses:

incredibly performant
fault-tolerant

Riak

Riak is a distributed key-value database where values can be anything—from plain text, JSON, or XML to images or video clips—all accessible through a simple HTTP interface. Whatever data you have, Riak can store it.

Riak is also fault-tolerant. Servers can go up or down at any moment with no single point of failure. As core developer Justin Sheehy once noted, “[The Riak team] focused so hard on things like write availability...to go back to sleep.”
Doesn't have foreign key.

Riak is a great choice for datacenters like Amazon that must serve many requests with low latency. It’s easy to manage, easy to set up, and can grow with your needs. If you’ve ever used Amazon Web Services, like SimpleDB or S3, you may notice some similarities in form and function. This is no coincidence. Riak is inspired by Amazon’s Dynamo paper.

curl -v -X PUT http://localhost:8091/riak/favs/db \
  -H "Content-Type: text/html" \
  -d "<html><body><h1>My new favorite DB is RIAK</h1></body></html>"

PUT something

If you navigate to http://localhost:8091/riak/favs/db in a browser, you’ll get a nice message from yourself.

Riak is a key-value store, so it expects you to pass in a key to retrieve a value. Riak breaks up classes of keys into buckets to avoid key collisions

http://SERVER:PORT/riak/BUCKET/KEY

putting a first value into a bucket name will create that bucket

$ curl -v -X PUT http://localhost:8091/riak/animals/ace \
  -H "Content-Type: application/json" \
  -d '{"nickname" : "The Wonder Dog", "breed" : "German Shepherd"}'

First released in 2009, Redis (REmote DIctionary Service) is a simple-to-use key-value store with a sophisticated set of commands. And when it comes to speed, Redis is hard to beat. Reads are fast, and writes are even faster, han- dling upwards of 100,000 SET operations per second by some benchmarks.

Redis

$ redis-server

$ redis-cli
redis 127.0.0.1:6379> PING
PONG

redis 127.0.0.1:6379> SET 7wks http://www.sevenweeks.org/
OK
redis 127.0.0.1:6379> GET 7wks
"http://www.sevenweeks.org/"

Columnar

Columnar, or column-oriented, databases are so named because the important aspect of their design is that data from a given column (in the two-dimensional table sense) is stored together. By contrast, a row-oriented database (like an RDBMS) keeps information about a row together.

In column- oriented databases, adding columns is quite inexpensive and is done on a row-by-row basis. Each row can have a different set of columns, or none at all, allowing tables to remain sparse without incurring a storage cost for null values. With respect to structure, columnar is about midway between relational and key-value.

RowId	EmpId	LastName	FirstName	Salary
001	10	Smith	Joe	40000
002	12	Jones	Mary	50000
003	11	Johnson	Cathy	44000
004	22	Jones	Bob	55000

Row-oriented

001:10,Smith,Joe,40000;
002:12,Jones,Mary,50000;
003:11,Johnson,Cathy,44000;
004:22,Jones,Bob,55000;

Column-oriented

10:001,12:002,11:003,22:004;
Smith:001,Jones:002,Johnson:003,Jones:004;
Joe:001,Mary:002,Cathy:003,Bob:004;
40000:001,50000:002,44000:003,55000:004;

HBase

HBase is built on Hadoop (a mapreduce engine) and designed for scaling horizontally on clusters of commodity hardware. HBase makes strong consistency guarantees and features tables with rows and columns—which should make SQL fans feel right at home. Out-of-the-box support for versioning and compression sets this database apart in the “Big Data” space.

Document-oriented

A document is like a hash, with a unique ID field and values that may be any of a variety of types, including more hashes.

Documents can contain nested structures, and so they exhibit a high degree of flexibility, allowing for variable domains.

MongoDB

MongoDB is designed to be huge (the name mongo is extracted from the word humongous). Mongo server configurations attempt to remain consistent—if you write something, subsequent reads will receive the same value (until the next update).

$ mongo book

> db.towns.insert({
name: "New York",
population: 22200000,
last_census: ISODate("2009-07-31"), famous_for: [ "statue of liberty", "food" ], mayor : {
name : "Michael Bloomberg",
party : "I" }
})

> db.towns.find()
{
	"_id" : ObjectId("4d0ad975bb30773266f39fe3"),
	"name" : "New York",
	"population": 22200000,
	"last_census": "Fri Jul 31 2009 00:00:00 GMT-0700 (PDT)", 
	"famous_for" : [ "statue of liberty", "food" ],
	"mayor" : { "name" : "Michael Bloomberg", "party" : "I" } 
}

db.towns.find({ "_id" : ObjectId("4d0ada1fbb30773266f39fe4") }) 
{
	"_id" : ObjectId("4d0ada1fbb30773266f39fe4"),
	"name" : "Punxsutawney",
	"population" : 6200,
	"last_census" : "Thu Jan 31 2008 00:00:00 GMT-0800 (PST)", 
	"famous_for" : [ "phil the groundhog" ],
	"mayor" : { "name" : "Jim Wehrle" } 
}

CouchDB

CouchDB targets a wide variety of deployment scenarios, from the datacenter to the desktop, on down to the smartphone. Written in Erlang, CouchDB has a distinct ruggedness largely lacking in other databases. With nearly incorruptible data files, CouchDB remains highly available even in the face of intermittent connectivity loss or hardware failure.

$ curl http://localhost:5984/music/74c7a8d2a8548c8b97da748f43000ac4 {
  "_id":"74c7a8d2a8548c8b97da748f43000ac4",
  "_rev":"4-93a101178ba65f61ed39e60d70c9fd97",
  "name":"The Beatles",
  "albums": [
    {
      "title":"Help!",
      "year":1965
    },{
      "title":"Sgt. Pepper's Lonely Hearts Club Band",
      "year":1967
    },{
      "title":"Abbey Road",
      "year":1969
} ]
}

Graph

One of the less commonly used database styles, graph databases excel at dealing with highly interconnected data. A graph database consists of nodes and relationships between nodes. Both nodes and relationships can have properties—key-value pairs—that store data. The real strength of graph databases is traversing through the nodes by following relationships.

Neo4J

One operation where other databases often fall flat is crawling through self-referential or otherwise intricately linked data. This is exactly where Neo4J shines. The benefit of using a graph database is the ability to quickly traverse nodes and relationships to find relevant data. Often found in social networking applications, graph databases are gaining traction for their flexibility, with Neo4j as a pinnacle implementation.

Alice has a bit of a sweet tooth and so is a big ice wine fan.

alice = g.addVertex([name:'Alice'])
ice_wine = g.V.filter{it.name=='Prancing Wolf Ice Wine 2007'}.next()
g.addEdge(alice, ice_wine, 'likes')

Tom loves Kabinett and ice wine and trusts anything written by Wine Expert Monthly.

tom = g.addVertex([name:'Tom'])
kabinett = g.V.filter{it.name=='Prancing Wolf Kabinett 2002'}.next()
g.addEdge(tom, kabinett, 'likes')
g.addEdge(tom, ice_wine, 'likes')
g.addEdge(tom, g.V.filter{it.name=='Wine Expert Monthly'}.next(), 'trusts')

Patty is friends with both Tom and Alice but is new to wine and has yet to choose any favorites.

patty = g.addVertex([name:'Patty'])
g.addEdge(patty, tom, 'friends')
g.addEdge(patty, alice, 'friends')

Without changing any fundamental structure of our existing graph, we were able to superimpose behavior beyond our original intent. The new nodes are related, as visualized in the following:

NoSQL Databases

By Sania Sutula

No relational Databases

Central question

The question you must always ask yourself

Types

Key-Value

Riak

PUT something

Redis

Redis

Columnar

Row-oriented

Column-oriented

HBase

Document-oriented

MongoDB

CouchDB

Graph

Neo4J

NoSQL Databases

More from Sania Sutula