318
N. Balac
6.1.3 Velocity
Velocity refers to the speed at which data is being transacted. Streaming data can
arrive in milliseconds and require a response within seconds or less. For example,
Facebook’s data warehouse not only stores hundreds of petabytes of data but needs
to accommodate data coming in at the rate of more than 600 terabytes per data
per day. Similarly, Google processes more than 3.5 billion searches per day, which
translates to a velocity of over 40,000 search queries per second.
6.1.4 Variety
The huge variety of the types and structures of data has become one of the critical
challenges of Big Data. Big Data requires systems to handle not only structured data
but also semi-structured and unstructured data as well. As described above, most
Big Data is in unstructured forms such as audio, images, video files, social media
updates, log files, click data, machine and sensor data, etc. Unfortunately, most
analytics techniques have traditionally been focused on analyzing only structured
data, so new techniques and approaches need to be developed. This fact alone helps
explain why there are such a large growing number of start-ups in Big Data.
6.1.5 Veracity
Perhaps the most nuanced of the four Vs is veracity. Veracity describes how
accurate and trustworthy data is in predicting business value through Big Data
analytics. Uncertainty in data is typically due to inconsistency, incompleteness,
latency ambiguities, approximations, etc. Data must be able to be verified based
on both accuracy and context. It is necessary to identify the right amount of highquality data that can be analyzed in order to impact business or outcomes.
These four definitions suggest that Big Data typically requires resources (computation and data infrastructure, tools, techniques, expertise, etc.) beyond the
current capabilities of many organizations [1]. Big Data solutions comprise a set
of analytical tools that are geared toward the fast and meaningful processing of
large data sets. This is a key aspect of Big Data analytics, as the goal is to derive
meaning or insight from data that can be used for making data-driven business
decisions. Big Data can be thought of as a process that is used when traditional
data processing and handling techniques alone cannot uncover the insights and
meaning of the underlying data. Often times, real-time processing is needed for a
massive amount of different types and frequencies of data in order to reveal patterns,
trends, and associations. This is especially true when relating to human behavior
and interactions. The goal for Big Data is to enable organizations to gather, store,
N. Balac
6.1.3 Velocity
Velocity refers to the speed at which data is being transacted. Streaming data can
arrive in milliseconds and require a response within seconds or less. For example,
Facebook’s data warehouse not only stores hundreds of petabytes of data but needs
to accommodate data coming in at the rate of more than 600 terabytes per data
per day. Similarly, Google processes more than 3.5 billion searches per day, which
translates to a velocity of over 40,000 search queries per second.
6.1.4 Variety
The huge variety of the types and structures of data has become one of the critical
challenges of Big Data. Big Data requires systems to handle not only structured data
but also semi-structured and unstructured data as well. As described above, most
Big Data is in unstructured forms such as audio, images, video files, social media
updates, log files, click data, machine and sensor data, etc. Unfortunately, most
analytics techniques have traditionally been focused on analyzing only structured
data, so new techniques and approaches need to be developed. This fact alone helps
explain why there are such a large growing number of start-ups in Big Data.
6.1.5 Veracity
Perhaps the most nuanced of the four Vs is veracity. Veracity describes how
accurate and trustworthy data is in predicting business value through Big Data
analytics. Uncertainty in data is typically due to inconsistency, incompleteness,
latency ambiguities, approximations, etc. Data must be able to be verified based
on both accuracy and context. It is necessary to identify the right amount of highquality data that can be analyzed in order to impact business or outcomes.
These four definitions suggest that Big Data typically requires resources (computation and data infrastructure, tools, techniques, expertise, etc.) beyond the
current capabilities of many organizations [1]. Big Data solutions comprise a set
of analytical tools that are geared toward the fast and meaningful processing of
large data sets. This is a key aspect of Big Data analytics, as the goal is to derive
meaning or insight from data that can be used for making data-driven business
decisions. Big Data can be thought of as a process that is used when traditional
data processing and handling techniques alone cannot uncover the insights and
meaning of the underlying data. Often times, real-time processing is needed for a
massive amount of different types and frequencies of data in order to reveal patterns,
trends, and associations. This is especially true when relating to human behavior
and interactions. The goal for Big Data is to enable organizations to gather, store,