Extending Standard MapReduce Algorithms

The information rate nowadays is expanding very quickly and contains complex and heterogeneous data types (text, images, videos, GPS data, purchase transactions) that require powerful computing engines, able to easily store and process such complex structures. Gartner's definition of the 3Vs (v...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	2016 IEEE Second International Conference on Multimedia Big Data (BigMM) s. 159 - 165
Hlavní autoři:	Hashem, Hadi, Ranc, Daniel
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	IEEE 01.04.2016
Témata:	Algorithm design and analysis BigData Tools Computational modeling Data Modeling Data models Data processing Engines MapReduce Algorithms Pre-Processing Sorting
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	The information rate nowadays is expanding very quickly and contains complex and heterogeneous data types (text, images, videos, GPS data, purchase transactions) that require powerful computing engines, able to easily store and process such complex structures. Gartner's definition of the 3Vs (volume, velocity, variety) describing this expansion of data will then lead to extract the unnamed forth V (value) from BigData. This added value addresses the need for valuation of enterprise data. In this paper, we discuss the existing MapReduce implementation techniques and the need of a different approach based on the pre-processing of the data. The goal is to show interesting results in terms of data processing costs, performance and green computing.
DOI:	10.1109/BigMM.2016.49