Database Reference
In-Depth Information
Traditionally, all of these problems would use graph object models (implemented by the
development team), and store the data in normalized form in a set of tables with foreign
key associations.
In the previous section you saw how a social network can be stored in the relational model
using two tables and a foreign key constraint. Using a Neo4j graph database, you can pre-
servethenaturalgraphstructureofthedata,withoutcompromisingperformanceorstorage
space, therefore improving the query performance significantly.
When using a Neo4j graph database to solve graph problems, the amount of work for the
developer is reduced, as the graph structures are already available via the database APIs.
Thedataisstoredinitsnaturalformat—asgraphnodesandrelationships—sotheeffectsof
an object-relational mismatch are minimized. And the graph query operations are executed
inanefficientandperformantwaybyusingtheNeo4jTraversalAPI.Throughoutthisbook
we'll demonstrate the power of Neo4j by solving problems such as access control lists and
social networks with recommendation engines.
If you're not familiar with graph theory, don't worry; we'll introduce some of the graph
theory concepts and algorithms, as well as their applicability to graph storage engines such
as Neo4j, when they become relevant in later chapters.
In the next section we'll discuss the reasons why graph databases (and other NoSQL tech-
nologies) started to gain popularity in the computer industry. We'll also discuss categories
of NoSQL systems with a focus on their differences and applicability.
1.6. Neo4j in NoSQL space
Since the beginning of computer software, the data that applications have had to deal with
hasgrownenormouslyincomplexity.Thecomplexityofthedataincludesnotonlyitssize,
but also its interconnectedness, its ever-changing structure, and concurrent access to the
data.
With all these aspects of changing data, it has been recognized that relational databases,
which have for a long time been the de facto standard for data storage, are not the best fit
forallproblemsthatincreasingly complexdatarequires.Asaresult,anumberofnewstor-
age technologies has been created, with a common goal of solving the problems relational
