Page 215 - Big data - Concept and application for telecommunications

Big data - Concept and application for telecommunications 5

From the IoT data provider point of view, the IoT needs to cope with a rapidly increasing number and types
of devices and needs to provide IoT data in an efficient way to facilitate data operation execution and data
consumption.
From the IoT data carrier point of view, the IoT needs to deal with a large number of devices accessing the
network and transferring large amounts of data. These IoT data, related to different IoT applications, have
different service level agreement (SLA) and quality of service (QoS) requirements: e.g., some IoT data have
real time QoS requirements (for example, real-time monitoring IoT data and emergency alarm IoT data),
others have no strict QoS requirements (for example, IoT data from electricity meters). These aspects raise
challenges in terms of IoT data transfer handling, including the avoidance of possible network congestions
from concurrent access of devices and gateways and in terms of active scheduling data transmission (e.g.,
among IoT application servers and IoT platforms).
From the IoT data framework provider point of view, the IoT needs to leverage general IoT data processing
capabilities and related infrastructure, in order to be accessible via open and standard interfaces, provide
APIs to the IoT data application provider, interoperate with other IoT data processing capabilities and with
data processing related IoT applications, while providing integrity, privacy and security protection for IoT
data.
From the IoT data application provider point of view, the IoT needs to be scalable and flexible for support of
applications based on IoT data processing capabilities, enabling the processing of variable sizes of data sets,
differentiated speed of data streams in and out of the IoT application, diversity of data formats to be
processed by the IoT application while providing integrity, privacy and security protection for IoT data.
From the IoT data consumer point of view, the IoT needs to provide consumers with the desired data
according to their requirements, including quality of experience (QoE).
Besides, collection, transfer, processing and consumption in a secure way and assurance of appropriate
privacy of IoT data, including personal data, are challenges which concern all IoT data roles.

7 Requirements of the IoT for big data
This clause complements the common requirements of the IoT [ITU-T Y.4100] in terms of the specific
requirements that the IoT is expected to support in order to address the challenges related to big data.

7.1 IoT data operations in the IoT for big data

Big data in the IoT are IoT data which have characteristics of high volume, high variety, high velocity and high
volatility as described in clause 6.1 and are hard to operate using existing tools and methods. To handle
challenges faced by big data in the IoT as described in clause 6.3, requirements of the IoT for big data are
specified for the IoT components, i.e., device, gateway, network, IoT platform and IoT application server as
described in [ITU-T Y.4113], from the view point of IoT data operations as shown in Figure 3.

Considering that the diverse set of concrete IoT deployments do not imply a unique logical sequencing of the
various IoT data operations, Figure 3 provides an abstract representation of the various IoT data operations
and related data flows.

Network and infrastructure 207