Page 64 - Big Data Analytics for Connected Vehicles and Smart Cities

44 Big Data Analytics for Connected Vehicles and Smart Cities What Is Big Data? 45

Sharing

Data sharing has the potential to be one of the biggest challenges facing trans-
portation agencies. The physical ability to share data over telecommunications
networks utilizing fiber optics and copper wire is just the beginning. It has been
learned that unless there is also a data-sharing agreement in place then noth-
ing will be transmitted across the telecommunications network. Data sharing
can also be a challenge between the public sector and the private sector. The
definition of suitable sharing agreements is paramount to enabling data flow. To
effectively address data sharing issues, it is also necessary to develop a business
model for the data. One business model could be that the public sector simply
collects the data as part of its everyday operations to deliver effective transporta-
tion and then provides this data free to private sector organizations that wish
to make it the basis for their products and services. Under another approach,
public sector agencies would attempt to develop agreements to enable them
to share in the value of the products and services that are derived from public
sector data. The definition of a business model for data would also include the
detailed definition of how much data processing is conducted within a public
agency and the cost of data collection and processing.
Storage

The data storage challenge revolves around the management of data sets across
data storage infrastructure. Unless these infrastructures are properly managed
and structured, it is possible to spend an undue amount of time and money on
data storage. This takes us back to the distinction between real-time data and
archive data. One of the obvious ways to address the structure of data storage
would be to separate them so that the more expensive real-time ways to store
and access data are used on the appropriate data set. It is interesting to note
that this problem is being driven by an extremely positive development in the
market—the cost of acquiring data storage has been dropping dramatically.
However, this has led to hidden costs beyond acquisition when it comes to
maintaining and managing the data. Figure 3.7 illustrates the reduction in the
cost to acquire data storage, showing the average hard drive cost per gigabyte
from 1980 to 2009 [10].
Over time, the cost of hard drive storage acquisition has dropped signifi-
cantly from a peak value $700,000 per gigabyte to a low of $0.3 per gigabyte.
Of course, acquiring the hard disk space is only one element in the cost of
data storage. Other elements include maintenance, upgrades, power, physical
facilities for hosting the hard drive storage, and the cost of managing it. How-
ever, the cost of hard drive storage acquisition seems to be a good yardstick to
indicate how dramatic the cost reduction has been. Another option that has
emerged in recent times involves simply renting or leasing data storage space at