Page 359 - Big data - Concept and application for telecommunications
P. 359
Big data - Concept and application for telecommunications 7
defined data model and are not organized in any defined manner. Within all data types data can exist in
formats, such as text, spreadsheet, video, audio, image, map, etc. [ITU-T Y.3600].
Big data is used in many fields, where data processing is characterized by scale (volume), diversity (variety),
speed (velocity) and possibly others like credibility (veracity) or business value, if traditional methods and
tools are not efficient. These characteristics, usually called v's, can be explained as following [ITU-T Y.3600]:
− Volume: refers to the amount of data collected, stored, analyzed and visualized, which big data
technologies need to resolve;
− Variety: refers to different data types and data formats that are processed by big data technologies;
− Velocity: refers to both how fast the data is collected and how fast the data is processed by big data
technologies to deliver expected results.
NOTE – Additionally, veracity refers to the uncertainty of data, and value refers to the business results from
gaining new information using big data technologies. Other v's can be considered as well.
Taking into account the described above v's characteristics, big data technologies and services can resolve
many new challenges, and can also create more new opportunities than ever before [ITU-T Y.3600]:
− Heterogeneity and incompleteness: data processed using big data can miss some attributes or
introduce noise into data transmission. Even after data cleaning and error correction, some
incompleteness and some errors in data are likely to remain. These challenges can be managed
during data analysis [b-CRA].
− Scale: processing of large and rapidly increasing volumes of data is a challenging task. Using data
processing technologies, the data scale challenge is mitigated by evolution of processing and storage
resources. However, nowadays data volumes are scaling faster than resources are evolving.
Technologies such as parallel databases, in-memory databases, non-SQL databases and analytical
algorithms resolve this challenge.
− Timeliness: the acquisition rate and timeliness, to effectively find elements in a limited-time period
that meet a specified criterion in a large dataset, are new challenges faced by data processing. Other
new challenges are related to the types of criteria specified, and need to devise new index structures
and responses to the queries having tight response-time limits.
− Privacy: data about human individuals, such as: demographic information, Internet activities,
commutation patterns, social interactions, energy or water consumption, are being collected and
analyzed for different purposes. Big data technologies and services are challenged to protect
personal identities and sensitive attributes of data throughout the entire data processing process,
while respecting applicable data retention policies.
Positive resolution of the above challenges opens new opportunities to discover new data relationships,
hidden patterns or unknown dependencies [ITU-T Y.3600].
6.2 Benefits of big data
Big data technologies can provide many benefits such as data accessibility, productivity of business processes,
and cost reduction to private via public sector.
Big data technology increases data accessibility by:
− Unlocking significant value by making information transparent;
− Creating and storing transactional data in digital form;
− Reducing time for finding/accessing the correct data.
Big data technology improves productivity by:
− Real-time monitoring and forecasting of events that impact either business performance or
operations;
− Timely insights from the vast amount of data;
− Identifying significant information that can improve decision quality or minimize risks;
Standardization efforts at a glance – roadmap 351