Page 358 - Big data - Concept and application for telecommunications

7 Big data - Concept and application for telecommunications

ICT Information and Communications Technology
IEC International Engineering Consortium

IoT Internet of Things
ISO International Organization for Standardization

JSON Java Script Object Notation
JTC 1 Joint Technical Committee 1
KVDB Key-Value Database Application Interface

LDP Linked Data Platform
M2M Machine to Machine

MQTT Message Queuing Telemetry Transport
PMML Predictive Model Markup Language

RDF Resource Description Framework
SC Subcommittee

SDN Software-defined Networking
SDO Standards Development Organization
SG Study Group

TC Technical Committee
URL Uniform Resource Locator

W3C World Wide Web Consortium
WG Working Group
XMILE XML Interchange Language

XML Extensible Markup Language

5 Conventions
None.

6 Landscape of big data from an ITU-T perspective

6.1 Characteristics and general concepts of big data
[ITU-T Y.3600] describes the characteristics and general concepts of the big data ecosystem.

With the rapid development of information and communications technology (ICT), Internet technologies and
services, huge amount of data are generated, transmitted and stored with explosive growth. Data are
generated by many sources and not only sensors, cameras, network devices, web pages, email systems, social
networks and many other sources. Datasets are becoming so large and complex or are arriving so fast that
traditional data processing methods and tools are inadequate. Efficient analytics of data within tolerable
elapsed times becomes very challenging. The paradigm being developed to resolve the above issues are
called big data [ITU-T Y.3600].
Within big data ecosystem, data types include structured, semi-structured and unstructured data. Structured
data are often stored in databases which may be organized in different models, such as relational model,
document model, key-value model, graph model etc. Semi-structured data does not conform to the formal
structure of data models, but contain tags or markers to identify data. Unstructured data do not have a pre-

350 Standardization efforts at a glance – roadmap