Geography Reference
In-Depth Information
Table 18.1
Contemporary technologies for supporting the life cycle of big data
Techs\BD processing
Collection
Storage
Management
Processing
Simulation
visualization
Cloud computing
x
x
x
x
Spatiotemporal
x
x
x
x
x
x
Data mining
x
x
x
Crowd-sourcing/grid
computing
x
x
GPU computing
x
volume and unstructured system during the lifecycle of data collection/generation,
storage/archive, discovery, processing/modeling, analyses, and visualization. Con-
temporary computing technologies (such as grid computing, cloud computing, and
GPU computing) and advanced methodologies (such as spatiotemporal statistics
and spatiotemporal data visualization) provide new tools and software to enable
handling big data. This chapter uses several projects as examples to illustrate how
big data are handled with the computing and advanced technologies. Table
18.1
summarizes how different technologies are used to support the lifecycle of big
data handling. The boxes checked means that the technology is used to address
the challenges within relevant data handling stage. For example, cloud computing
can be used to help manage storage for convenient storage access.
While the processing of big data starts as a new research and in the development
direction, many new issues remain to be researched. To name a few: security,
large/fast data storage, sufficient computing power, and smart and fast data pro-
cessing.
Security is a big concern for both sensitive data and the information systems that
handles the big data when the data is archived across the global environment
and shared among global users. Relevant policies and regulations should be
developed to address this problem.
Storage is another concern in terms of both volume and speed of storage as well
as the manageability of storage for each data access including writing, reading,
selecting, and updating.
Computing power is short for big data handling. How to reduce the comput-
ing requirements and best leverage different types of computing resources in
combination, such as using both cloud computing and grid computing for data
processing.
Smart and fast data processing is critical for real time applications, such as
emergency response and decision relevant event detection.
Acknowledgements
Research reported is supported by Microsoft Research, NSF (IIP-1160979),
and NASA (NNX12AF89G).
Search WWH ::
Custom Search