Global Positioning System Reference
In-Depth Information
it is necessary to acquire powerful servers, pay the salaries of database
administrators, and invest in caching technologies such as Memcached. 1
As a result, new systems that can support such requirements have been
proposed. An implicit requirement of these systems is to scale to a large
number of nodes without degrading the quality of the services they deliver
and with a low budget.
These systems were called NoSQL systems (or data stores). Currently,
they are becoming increasingly popular among Web companies. The NoSQL
systems represent an alternative solution for the needs of modern interactive
software systems. They are designed to scale to thousands or millions of users
performing simple data operations such as updates and reads. Proponents
of NoSQL systems state that these systems provide simpler scalability
and improved performance in contrast to traditional relational database
systems. These are crucial features for Web 2.0 companies such as Facebook,
Amazon, and Google. In addition, NoSQL systems excel at storing large
amounts of unstructured data known as Big Data (Floratou et al. 2012). A recent
report released by the Gartner team identifi es NoSQL systems as one the top 10
technology trends that will have impact on information management (Gartner
2013). Some examples of NoSQL systems include MongoDB, CouchDB,
BigTable, Cassandra, and Neo4j.
In principle, NoSQL systems were proposed to manage large amounts
of non-spatial data. However, the spatial dimension has proven to be of
particular interest in Web 2.0 applications (Schutzberg 2011). This can
be observed with the emergence of location-based applications (e.g.,
Foursquare) 2 and others that allow users to geocode shared content (e.g.,
Twitter, 3 Facebook, 4 Flickr, 5 and Panoramio 6 ). All of these applications use
a NoSQL system in the background.
As a result, NoSQL systems began to allow the storage and retrieval
of spatial data (Baas 2012). Some NoSQL systems include support for
geospatial data either natively or with an extension. Others were not
originally designed for geospatial applications but have been extended to
support geospatial data. Examples of NoSQL systems that provide support
for spatial data include: CouchDB, MongoDB, BigTable, and Neo4j. In this
chapter, we have chosen to compare these four NoSQL systems, as they
represent the main players that support spatial data. In the past, Cassandra
coped with the spatial dimension, but that is not the case anymore.
1 Memcached, http://memcached.org/
2 Foursquare, http://foursquare.com/
3 Twitter, http://twitter.com/
4 Facebook, http://www.facebook.com/
5 Flickr, http://www.fl ickr.com/
6 Panoramio, http://www.panoramio.com/
Search WWH ::




Custom Search