Page 19 - Big data - Concept and application for telecommunications

Big data - Concept and application for telecommunications 1

6.2.1.2 Data broker

The data broker serves to connect between the data supplier and the big data service provider. The data
broker can act as a clearinghouse, open data mart, etc. and its activities include:
– providing a meta-information registry to the data supplier for publishing their data sources;
– finding on-line open-data sources and registering corresponding meta-information;
– providing a service catalogue to the big data service provider for searching usable data.

6.2.2 Big data service provider (BDSP)
The big data service provider (BDSP) supports capabilities for big data analytics and infrastructure. The big
data service provider can act as a form of big data platform, an extension of the existing data analytics
platform, etc. Big data service provider activities include:

– searching data sources (from the data broker) and collecting data by requesting and crawling;
– storing data to a data repository;
– integrating data;
– providing tools for data analysis and visualization;
– supporting data management such as data provenance, data privacy, data security, data retention
policy, data ownership, etc.

6.2.3 Big data service customer (BDC)
The big data service customer (BDC) is the end-user or is a system that uses the results or services from a big
data service provider. The big data service customer may produce new services or knowledge on consumer
activities and furnish them outside of the big data ecosystem. Big data service customer activities include:
– requesting big data services from the big data service provider;
– using the outputs of big data services.

6.3 Relationship between cloud computing and big data

Big data refers to technologies and services which extract valuable information from the extensive datasets
characterized by the Vs, while cloud computing is, as defined in [ITU-T Y.3500], the paradigm for enabling
network access to a scalable and elastic pool of shareable physical or virtual resources with self-service
provisioning and administration on-demand.
Big data needs on-demand high performance data processing and distributed storage as well as variety of
tools required to accomplish activities of the big data ecosystem which are described in clause 6.2. Cloud
computing meets the challenges of big data as described in clause 6.1. The burst nature of workloads makes
cloud computing more appropriate for big data challenges such as scalability and timeliness. The big data
ecosystems, which are supported by a cloud computing system context, can be referred to as cloud
computing based big data. Cloud computing based big data is addressed in detail in clause 7.

7 Cloud computing based big data
This clause describes a cloud computing based big data system context that is effective for supporting big
data. It also provides benefits of cloud computing based big data.

7.1 Cloud computing based big data system context
Cloud computing based big data system context is described with new sub-roles and activities based on the
architectural user view defined in [ITU-T Y.3502]. This clause describes how cloud computing can support the
three main roles in a big data ecosystem: data provider, big data service provider and big data service
customer.

Cloud computing sub-roles can be mapped to big data roles as shown in Table 7-1.

Basics of Big data 11