Database Reference
In-Depth Information
methods. The performance can only be evaluated by the system is implemented
and deployed, which could not horizontally compare advantages and disadvan-
tages of various solutions and compare efficiencies before and after the use of big
data. In addition, since data quality is an important basis of data preprocessing,
simplification, and screening, it is also an urgent problem to effectively evaluate
data quality.
Evolution of Big Data Computing Modes
: This includes external storage mode,
data flow mode, PRAM mode, and MR mode, etc. The emergence of big data
triggers the development of algorithm design, which has transformed from a
computing-intensive approach into a data-intensive approach. Data transfer has
been a main bottleneck of big data computing. Therefore, many new computing
models tailored for big data have emerged and more such models are on the
horizon.
7.1.2
Technology Development
The big data technology is still in its infancy. Many key technical problems, such as
cloud computing, grid computing, stream computing, parallel computing, big data
architecture, big data model, and software systems supporting big data, etc. should
be fully investigated.
Format Conversion
: Due to wide and various data sources, heterogeneity is
always a characteristic of big data, as well as a key factor which restricts the
efficiency of data format conversion. If such format conversion can be made more
efficient, the application of big data may create more values.
Big Data Transfer
: Big data transfer involves big data generation, acquisition,
transmission, storage, and other data transformations in the spatial domain.
As discussed, big data transfer usually incurs high costs, which is also the
bottleneck for big data computing. However, data transfer is inevitable in big
data applications. Improving the transfer efficiency of big data is a key factor to
improve big data computing.
Real-time Performance
: The real-time performance of big data is also a core
problem in many different application scenarios. Ways to define the life cycle
of data, compute the rate of depreciation of data, and build computing models
of real-time applications and online applications, will influence the values and
analytical and feedback results of big data.
As big data research is advanced, new problems on big data processing arise
from the traditional simple data analysis, including: (a) data re-utilization, since big
data features big value but low density, with the increase of data scale, more values
may be mined from re-utilization of existing data; (b) data re-organization, datasets
in different businesses can be re-organized, with the total re-organized data values
larger than the total datasets' value; (c) data exhaust, unstructured information or
data that is a by-product of the online activities of Internet users. In big data, not
Search WWH ::
Custom Search