nosql概述（引自维基百科nosql）

A NoSQL database provides a mechanism for storage and retrieval of data that is modeled in means other than the tabular relations used in relational databases. Motivations for this approach include simplicity of design, horizontal scaling and finer control over availability. NoSQL databases are often highly optimizedkey–value stores intended primarily for simple retrieval and appending operations, whereas an RDBMS is intended as a general purpose data store. There will thus be some operations where NoSQL is faster and some where an RDBMS is faster. NoSQL databases are finding significant and growing industry use in big dataand real-time web applications.^{[1] NoSQL systems are also referred to as "Not only SQL" to emphasize that they may in fact allow SQL-like query languages to be used. In the context of the CAP theorem, NoSQL stores often compromise consistency in favor of availability and partition tolerance. Barriers to the greater adoption of NoSQL data stores in practice include: the lack of full ACID transaction support, the use of low-level query languages, the lack of standardized interfaces, and the huge investments already made in SQL by enterprises. [2]}

 

 

History[edit]

Carlo Strozzi used the term NoSQL in 1998 to name his lightweight, open-source relational database that did not expose the standard SQL interface.^{[3] Strozzi suggests that, as the current NoSQL movement "departs from the relational model altogether; it should therefore have been called more appropriately 'NoREL'.[4]}

Eric Evans (then a Rackspace employee) reintroduced the term NoSQL in early 2009 when Johan Oskarsson of Last.fm wanted to organize an event to discuss open-source distributed databases.^{[5] The name attempted to label the emergence of a growing number of non-relational, distributed data stores that often did not attempt to provide atomicity, consistency, isolation and durability guarantees that are key attributes of classic relational database systems.[6]}

Taxonomy[edit]

There have been various approaches to classify NoSQL databases, each with different categories and subcategories. Because of the variety of approaches and overlaps it is difficult to get and maintain an overview of non-relational databases. Nevertheless, the basic classification that most would agree on is based on data model. A few of these and their prototypes are:

Column: HBase, Accumulo, Cassandra

Document: MarkLogic, MongoDB, Couchbase

Key-value: Dynamo, Riak, Redis, MemcacheDB, Project Voldemort

Graph: Neo4J, OrientDB, Allegro, Virtuoso

Classification based on data model[edit]

Stephen Yen in his blog post "NoSQL is a Horseless Carriage" suggests the following:^[7]

Term	Matching Database
KV Cache	Memcached, Repcached, Coherence, Infinispan, eXtreme Scale, JBoss Cache, Velocity, Terracotta, Gigaspaces XAP
KV Store	Keyspace, Flare, SchemaFree, RAMCloud
KV Store - Eventually consistent	Dynamo, Voldemort, Dynomite, SubRecord, MotionDb, DovetailDB
Data-structures server	Redis
KV Store - Ordered	TokyoTyrant, Lightcloud, NMDB, Luxio, MemcacheDB, Actord
Tuple Store	Gigaspaces, Coord, Apache River
Object Database	ZopeDB, DB4O, Shoal, Perst
Document Store	MarkLogic, CouchDB, MongoDB, Jackrabbit, XML-Databases, ThruDB, CloudKit, Persevere, Riak Basho, Scalaris
Wide Columnar Store	BigTable, HBase, Cassandra, Hypertable, KAI, OpenNeptune, Qbase, KDI

Classification based on feature[edit]

Ben Scofield categorized NoSQL databases based on nonfunctional categories (“(il)ities“) plus a rating of their feature coverage:^{[citation needed]}

Data Model	Performance	Scalability	Flexibility	Complexity	Functionality
Key–value Stores	high	high	high	low	variable (none)
Column Store	high	high	moderate	low	minimal
Document Store	high	variable (high)	high	low	variable (low)
Graph Database	variable	variable	high	high	graph theory
Relational Database	variable	variable	low	moderate	relational algebra.

It has been suggested that this article be merged into Comparison of structured storage software. (Discuss) Proposed since March 2011.

Examples[edit]

Document store[edit]

The central concept of a document store is the notion of a "document". While each document-oriented database implementation differs on the details of this definition, in general, they all assume that documents encapsulate and encode data (or information) in some standard formats or encodings. Encodings in use include XML, YAML, and JSON as well as binary forms like BSON, PDF and Microsoft Office documents (MS Word, Excel, and so on).

Different implementations offer different ways of organizing and/or grouping documents:

Collections

Tags

Non-visible Metadata

Directory hierarchies

Compared to relational databases, for example, collections could be considered as tables as well as documents could be considered as records. But they are different: every record in a table has the same sequence of fields, while documents in a collection may have fields that are completely different.

Documents are addressed in the database via a unique key that represents that document. One of the other defining characteristics of a document-oriented database is that, beyond the simple key-document (or key–value) lookup that you can use to retrieve a document, the database will offer an API or query language that will allow retrieval of documents based on their contents.

Name	Language	Notes
BaseX	Java, XQuery	XML database
Cloudant	Erlang, Java, Scala, C	JSON store (online service)
Clusterpoint	C++	XML, geared for Full text search
Couchbase Server	Erlang, C, C++	Support for JSON and binary documents
Apache CouchDB	Erlang	JSON database
djondb[8][9][10]	C++	JSON, ACID Document Store
ElasticSearch	Java	JSON, Search engine
eXist	Java, XQuery	XML database
Jackrabbit	Java	Java Content Repository implementation
IBM Lotus Notes and Lotus Domino	LotusScript, Java, IBM X Pages, others	MultiValue
MarkLogic Server	XQuery, Java, REST	XML database with support for JSON, text, and binaries
MongoDB	C++, C#, Go	BSON store (binary format JSON)
Oracle NoSQL Database	Java, C
OrientDB	Java	JSON, SQL support
CoreFoundation Property list	C, C++, Objective-C	JSON, XML, binary
Sedna	XQuery, C++	XML database
SimpleDB	Erlang	online service
TokuMX	C++, C#, Go	MongoDB with Fractal Tree indexing
OpenLink Virtuoso	C++, C#, Java, SPARQL	middleware and database engine hybrid

Graph[edit]

This kind of database is designed for data whose relations are well represented as a graph (elements interconnected with an undetermined number of relations between them). The kind of data could be social relations, public transport links, road maps or network topologies, for example.

Name	Language	Notes
AllegroGraph	SPARQL	RDF GraphStore
IBM DB2	SPARQL	RDF GraphStore added in DB2 10
DEX	Java, C++, .NET	High-performance graph database
FlockDB	Scala
InfiniteGraph	Java	High-performance, scalable, distributed graph database
Neo4j	Java
OpenLink Virtuoso	C++, C#, Java, SPARQL	middleware and database engine hybrid
OrientDB	Java
Sones GraphDB	C#
Sqrrl Enterprise	Java	Distributed, real-time graph database featuring cell-level security
OWLIM	Java, SPARQL 1.1	RDF graph store with reasoning

Key–value stores[edit]

Key–value stores allow the application to store its data in a schema-less way. The data could be stored in a datatype of a programming language or an object. Because of this, there is no need for a fixed data model.^{[11][12] The following types exist:}

KV - eventually consistent[edit]

Apache Cassandra

Dynamo

Hibari

OpenLink Virtuoso

Project Voldemort

Riak^[13]

KV - hierarchical[edit]

GT.M^[14]

InterSystems Caché

KV - cache in RAM[edit]

memcached

redis

OpenLink Virtuoso

Hazelcast

Oracle Coherence

KV - solid state or rotating disk[edit]

Aerospike

BigTable

CDB

Couchbase Server

Keyspace

LevelDB

MemcacheDB (using Berkeley DB)

MongoDB

OpenLink Virtuoso

Tarantool

Tokyo Cabinet

Tuple space

Oracle NoSQL Database

KV - ordered[edit]

Berkeley DB

FoundationDB

IBM Informix C-ISAM

InfinityDB

MemcacheDB

NDBM

Object database[edit]

db4o

GemStone/S

InterSystems Caché

JADE

NeoDatis ODB

ObjectDB

Objectivity/DB

ObjectStore

ODABA

Perst

OpenLink Virtuoso

Versant Object Database

WakandaDB

ZODB

Tabular[edit]

Apache Accumulo

BigTable

Apache Hbase

Hypertable

Mnesia

OpenLink Virtuoso

Tuple store[edit]

Apache River

OpenLink Virtuoso

Tarantool

Triple/Quad Store (RDF) database[edit]

SparkleDB

Virtuoso Universal Server

Ontotext-OWLIM

Apache JENA

Oracle NoSQL database

MarkLogic

Hosted[edit]

Freebase

OpenLink Virtuoso

Datastore on Google Appengine

Amazon DynamoDB

Cloudant Data Layer (CouchDB)

Multivalue databases[edit]

Northgate Information Solutions Reality, the original Pick/MV Database

Extensible Storage Engine (ESE/NT)

OpenQM

Revelation Software's OpenInsight

Rocket U2

D3 Pick database

InterSystems Caché

InfinityDB

Cell database[edit]

NoSQL databases on the cloud[edit]

NoSQL databases can be run on-premises, but are also often run on IaaS or PaaS platforms like Amazon Web Services, RackSpace or Heroku. There are three common deployment models for NoSQL on the cloud:

Virtual machine image - cloud platforms allow users to rent virtual machine instances for a limited time. It is possible to run a NoSQL database on these virtual machines. Users can upload their own machine image with a database installed on it, use ready-made machine images that already include an optimized installation of a database, or install the NoSQL database on a running machine instance.

Database as a service - some cloud platforms offer options for using familiar NoSQL database products as a service, such as MongoDB, Redis and Cassandra, without physically launching a virtual machine instance for the database. The database is provided as a managed service, meaning that application owners do not have to install and maintain the database on their own, and pay according to usage. Some database as a service providers provide additional features, such as clustering or high availability, that are not available in the on-premise version of the database (see the table below for several examples).

Native cloud NoSQL databases - some providers offer a NoSQL database service which is available only on the cloud. A well-known example is Amazon’sSimpleDB, a simple NoSQL key-value store. SimpleDB cannot be installed on a local machine and cannot be used on any cloud platform except Amazon’s.

The following table provides notable examples of NoSQL databases available on the cloud in each of these deployment models:

Deployment Model	Database Technology	Provider	Cloud-Specific Features	Pricing Model
Virtual machine image	MongoDB	MongoDB - machine images for Amazon EC2[15] and Windows Azure[16]	None	Database and machine image - open source Amazon/Azure instances - pay per use
Virtual Machine Image	Redis	Redis - standard open source installation Script for installation on Amazon EC2 [17] Recommended installation on Windows Azure [18]	None	Database and machine image - open source Amazon/Azure instances - pay per use
Virtual machine image	Cassandra	Apache Cassandra - machine image for Amazon EC2[19]	None	Database and machine image - open source Amazon instances - pay per use
Database as a Service	MongoDB	Mongolab[20] - available on Amazon, Google, Joyent, Rackspace and Windows Azure	Managed service High availability Automatic failover Pre-configured clustering	Free up to 500MB (on disk)[21] Paid plans based on architecture and storage size
Database as a Service	Redis/Memcached	Amazon Web Services - ElastiCache[22]	Managed service Automatic healing of failed nodes Resilient system to prevent overloaded DBs Performance monitoring	Free for 750 hours on micro instance[23] Pay per use for machine utilization, no separate charge for data usage[24]
Database as a Service	Redis	RedisToGo[25] - available on Amazon EC2, RackSpace, Heroku, AppHarbor, Orchestra	Managed service Daily backups API enabling creation, deletion, or download of Redis instances	Free up to 5MB (memory) Paid plans based on memory usage
Database as a Service	Redis	Redis Cloud (Garantia Data)[26] - available on Amazon EC2, Windows Azure, Heroku, Cloud Foundry, OpenShift, AppFog, AppHarbor	Managed service Automatic scaling, unlimited Redis nodes High availability Built-in clustering	Free up to 25MB (memory)[27] Pay per use
Database as a Service	Cassandra	InstaClustr[28] - available on Amazon EC2, RackSpace, Windows Azure, Joyent, Google Compute Engine	Managed service Performance tuning Monitoring Automated backups DataStax OpsCenter for cluster admin	Paid plans based on disk storage, memory usage and CPU cores[29]
Native cloud NoSQL database	Amazon SimpleDB	Amazon Web Services	Managed service High availability Unlimited scale Data durability	Free for 750 hours on micro instance[30] Pay per use - separate charge for machine utilization and data usage[30]
Native cloud NoSQL database	Google App Engine Datastore[31]	Google	No planned downtime Atomic transactions High availability of reads and writes	Free with quota system limiting instance hours, storage and throughput[32] Pay per use based on instance hours, storage, throughput and other parameters
Native cloud NoSQL database	SalesForce Database.com[33]	SalesForce	Unlimited scale Access to SalesForce meta data Social API Support for mobile clients Multi-tenancy	Free up to 100K records and 50K transactions[34] Pay per use based on users, number of records and transactions

See also[edit]

CAP theorem

Comparison of object database management systems

Comparison of structured storage software

Faceted search

Triplestore

Distributed cache

References[edit]

Further reading[edit]

Pramod Sadalage and Martin Fowler (2012). NoSQL Distilled: A Brief Guide to the Emerging World of Polyglot Persistence. Addison-Wesley. ISBN 0-321-82662-0.

Christof Strauch (2012). "NoSQL Databases".

Moniruzzaman AB, Hossain SA (2013). "NoSQL Database: New Era of Databases for Big data Analytics - Classification, Characteristics and Comparison".

Kai Orend (2013). Analysis and Classification of NoSQL Databases and Evaluation of their Ability to Replace an Object-relational Persistence Layer.

Ganesh Krishnan, Sarang Kulkarni, Dharmesh Kirit Dadbhawala. "Method and system for versioned sharing, consolidating and reporting information".

Sugam Sharma. "A Brief Review on Modern NoSQL Data Models, Handling Big Data".

External links[edit]

Christoph Strauch. "NoSQL whitepaper". Hochschule der Medien, Stuttgart.

Stefan Edlich. "NoSQL database List".

Peter Neubauer (2010). "Graph Databases, NOSQL and Neo4j".

Sergey Bushik (2012). "A vendor-independent comparison of NoSQL databases: Cassandra, HBase, MongoDB, Riak". NetworkWorld.

v t e Database management systems Database models Database normalization Database storage Distributed DBMS Federated database system Referential integrity Relational algebra Relational calculus Relational database Relational DBMS Relational model Object-relational database Transaction processing Concepts Database ACID CRUD Null Candidate key Foreign key Primary key Superkey Surrogate key Armstrong's axioms Objects Relation  table column row View Transaction Log Trigger Index Stored procedure Cursor Partition Components Concurrency control Data dictionary JDBC XQJ ODBC Query language Query optimizer Query plan Functions Administration and automation Query optimization Replication Database products   Object-oriented  comparison Relational  comparison Document-oriented Graph database NoSQL NewSQL